Crops, Culture, and Contact in Prehistoric Cyprus 9781407312767, 9781407342436

Recent archaeobotanical results from early Neolithic sites on Cyprus have put the island in the forefront of debates on

227 108 43MB

English Pages [169] Year 2014

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Front Cover
Title Page
Copyright
ABSTRACT
Table of Contents
LIST OF FIGURES
LIST OF TABLES
LIST OF APPENDICES
ACKNOWLEDGEMENTS
CHAPTER 1. THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION
CHAPTER 2. ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT
CHAPTER 3. ARCHAEOBOTANICAL MATERIALS AND METHODS
CHAPTER 4. HISTORY OF CYPRIOT ARCHAEOBOTANY
CHAPTER 5. IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS
CHAPTER 6. COMPARATIVE ARCHAEOBOTANICAL RESULTS
CHAPTER 7. INTERPRETATIONS, CONCLUSIONS, AND SUGGESTIONS FOR FUTURE RESEARCH
APPENDICES
REFERENCES CITED
Recommend Papers

Crops, Culture, and Contact in Prehistoric Cyprus
 9781407312767, 9781407342436

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Crops, Culture, and Contact in Prehistoric Cyprus Leilani Lucas

BAR International Series 2639 2014

ISBN 9781407312767 paperback ISBN 9781407342436 e-format DOI https://doi.org/10.30861/9781407312767 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

ABSTRACT Recent archaeobotanical results from early Aceramic Neolithic sites on Cyprus (c. 8,500 cal. BC) have put the island in the forefront of debates on the spread of Near Eastern agriculture, with domestic crops appearing on the island shortly after they evolved. The archaeobotanical results from these early Cypriot Aceramic Neolithic sites changed conventional views regarding the Cypriot prehistoric economy, specifically the timing of the introduction of farming to the island. However, what happened after the introduction of agriculture to Cyprus has been less discussed. This thesis explores the role of new crop introductions, local agricultural developments, and intensification in subsequent economic and social developments on Cyprus corresponding with the island’s evidence of ongoing social transformations and changing off-island patterns of contacts. In addition to contributing to discussions on the origins and spread of Near Eastern agriculture, this thesis contributes to current archaeological debates on external contact and the influence of the broader Near East on the development of the island’s prehistoric economy. Further, the primary objective of this research is the comparative quantitative analysis of the Cypriot charred macro botanical record including archaeobotanical data from four recently excavated Cypriot sites, Krittou Marottou-‘Ais Yiorkis, KissonergaSkalia, Souskiou-Laona, and Prastion-Mesorotsos. This research is a chronological and regional analysis of the botanical record of Cyprus and a comparison of data from similarly dated sites in the Levantine mainland, Turkey, and Egypt.

i

LIST OF CONTENTS Abstract .............................................................................................................................................................................. ii List of Contents.................................................................................................................................................................iii List of figures, tables, and appendices ............................................................................................................................ iv Acknowledgements ........................................................................................................................................................... ix CHAPTER 1 THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION............... 1 1.1 Introduction……. .......................................................................................................................................................... 1 1.2 Research Aims and Objectives ...................................................................................................................................... 1 1.3 General review of agricultural origins and its spread to Cyprus ................................................................................... 1 1.4 Agricultural systems and change ................................................................................................................................... 2 1.5 Relationship of early agriculture to food culture and technologies ............................................................................... 4 1.6 Agriculture and the emergence of the Cypriot Bronze Age .......................................................................................... 5 1.7 Research Questions ....................................................................................................................................................... 6 1.8 Archaeobotanical Samples ............................................................................................................................................ 7 1.9 Organisation of the book ............................................................................................................................................... 7 CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT ...... 9 2.1 Introduction ................................................................................................................................................................... 9 2.2 Cyprus, the study area ................................................................................................................................................... 9 2.3 Archaeological complexes of Cyprus and the mainland Levant ................................................................................. 12 2.4 Summary of Subsistence on Prehistoric Cyprus .......................................................................................................... 19 CHAPTER 3 ARCHAEOBOTANICAL MATERIALS AND METHODS ..................................................................................... 24 3.1 Introduction ................................................................................................................................................................. 24 3.2 Sampling on site: an overview .................................................................................................................................... 24 3.3 Krittou Marottou-‘Ais Yiorkis .................................................................................................................................... 24 3.4 Prastion-Mesorotsos .................................................................................................................................................... 25 3.5 Souskiou-Laona ........................................................................................................................................................... 26 3.6 Kissonerga-Skalia ........................................................................................................................................................ 27 3.7 Retrieval of plant material ........................................................................................................................................... 29 3.8 Sorting of the light fraction ......................................................................................................................................... 29 3.9 Identification ............................................................................................................................................................... 29 3.10 Compilation of database ............................................................................................................................................ 29 3.11 Quantification of the Remains ................................................................................................................................... 29 3.12 Statistical Methods .................................................................................................................................................... 29 CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY ................................................................................................... 31 4.1 Introduction ................................................................................................................................................................. 31 4.2 Presentation of taxa ..................................................................................................................................................... 31 4.3 Chronology .................................................................................................................................................................. 31 4.4 Introduction of flotation techniques in Cyprus ............................................................................................................ 31 4.5 Current archaeobotany in Cyprus ................................................................................................................................ 33 4.6 Aceramic Neolithic...................................................................................................................................................... 33 4.7 Ceramic Neolithic........................................................................................................................................................ 36 4.8 Chalcolithic ................................................................................................................................................................. 37 4.9 Bronze Age Occupation .............................................................................................................................................. 38 4.10 Late Bronze Age and Beyond .................................................................................................................................... 39 4.11 Cypriot Archaeobotanical Summary ......................................................................................................................... 39

ii

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS ....... 48 5.1 Introduction ................................................................................................................................................................. 48 5.2 Krittou Marottou-‘Ais Yiorkis .................................................................................................................................... 48 5.3 Prastion-Mesorotsos ................................................................................................................................................... 50 5.4 Souskiou-Laona .......................................................................................................................................................... 51 5.5 Kissonerga-Skalia ...................................................................................................................................................... 52 5.6 Fruitful contexts in Cypriot Archaeobotany ................................................................................................................ 54 5.7 Conclusions ................................................................................................................................................................. 55 CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS ......................................................................................... 56 6.1 Introduction ................................................................................................................................................................. 56 6.2 Refining the dataset ..................................................................................................................................................... 56 6.3 Comparative analysis .................................................................................................................................................. 58 6.4 Cyprus Results ............................................................................................................................................................. 66 6.5 Conclusions ................................................................................................................................................................. 68 CHAPTER 7 INTERPRETATIONS, CONCLUSIONS, AND SUGGESTIONS FOR FUTURE RESEARCH ..................................... 70 7.1 Introduction ................................................................................................................................................................. 70 7.2 Regionalism in Near Eastern Archaeobotany .............................................................................................................. 70 7.3 The Cypriot Interaction Sphere during the Neolithic .................................................................................................. 71 7.4 The Cypriot Interaction Sphere during the Chalcolithic and Bronze Age ................................................................... 72 7.5 External Contact during the Chalcolithic and Bronze Age of Cyprus ......................................................................... 75 7.6 Conclusions ................................................................................................................................................................. 75 7.7 Suggestions for Future Research ................................................................................................................................. 75 APPENDICES ...................................................................................................................................................................... 78 REFERENCES CITED

................................................................................................................................................... 144

iii

LIST OF FIGURES CHAPTER 2 Figure 2.1 Map of Cyprus showing geology, topography and phytogeogrpahics zones ..................................................... 9 Figure 2.2 Map of Cyprus showing average rainfall distribution ...................................................................................... 21 Figure 2.4 Map of Cyprus showing the locations of sites discussed in text ...................................................................... 21 CHAPTER 3 Figure 3.1 Krittou Marottou-‘Ais Yiorkis, photographs of site .......................................................................................... 24 Figure 3.2 Photographs of artefacts recovered from Krittou Marottou-‘Ais Yiorkis ......................................................... 25 Figure 3.3 Photograph of circular structure from Krittou Marottou-‘Ais Yiorkis .............................................................. 25 Figure 3.4 Area V Plan of Prastion-Mesorotsos ................................................................................................................ 26 Figure 3.5 Photographs of features at Prastion-Mesorotsos .............................................................................................. 26 Figure 3.6 Souskiou-Laona, photograph of site ................................................................................................................ 26 Figure 3.7 Souskiou-Laona, photograph of circular structure ........................................................................................... 27 Figure 3.8 Souskiou-Laona site plan ................................................................................................................................. 27 Figure 3.9 Kissonerga-Skalia site plan .............................................................................................................................. 28 Figure 3.10 Kissonerga-Skalia plan of Plot 199 ................................................................................................................ 28 Figure 3.11 Kissonerga-Skalia Trench B photograph ....................................................................................................... 28 CHAPTER 4 Figure 4.1 Calibrated radiocarbon dates from sites with archaeobotanical data ............................................................... 32 Figure 4.2 Cumulative number of publications and number of sites in Cypriot archaeobotany........................................ 32 Figure 4.3 Bar chart of the total number of taxa................................................................................................................ 43 Figure 4.4 Bar chart of the number of sites per phase with the presence domesticated wheat .......................................... 43 Figure 4.5 Bar chart of the number of sites per phase with the presence domesticated barley ......................................... 44 Figure 4.6 Bar chart of the number of sites per phase with the presence of legumes ........................................................ 44 CHAPTER 5 Figure 5.1 Correlation between number of identifiable items and the number of cereals Krittou Marottou-‘Ais Yiorkis . 48 Figure 5.2 Bar chart of the flowering time of the different arable weeds from Krittou Marottou-‘Ais Yiorkis ................. 50 Figure 5.3 Bar chart of the flowering time of the different arable weeds from Kissonerga-Mylouthkia ........................... 50 Figure 5.4 Pie chart of the proportional representation per building at Souskiou-Laona .................................................. 51 Figure 5.5 Scatter gram plot, Souskiou-Laona .................................................................................................................. 51 Figure 5.6 Scatter gram plot, Souskiou-Laona .................................................................................................................. 52 Figure 5.7 Scatter gram plot, Kissonerga-Skalia ............................................................................................................... 52 Figure 5.8 Scatter gram plot, Kissonerga-Skalia ............................................................................................................... 52 Figure 5.9 Bar chart of the flowering time of the different arable weeds from Kissonerga-Skalia ................................... 53 Figure 5.10 Scatter gram plot of the number of taxa and volume, Cyprus and the mainland Levant ............................... 54 Figure 5.11 Scatter gram plot of the number of taxa and number of items from sites located in Cyprus and the mainland Levant ................................................................................................................................................................................ 54 CHAPTER 6 Figure 6.1 The data model ................................................................................................................................................. 56 Figure 6.2 Map of regions compared in Correspondence Analysis ................................................................................... 57 Figure 6.3 CA samples plot of taxa, Neolithic .................................................................................................................. 58 Figure 6.4 CA bi-plot of taxa, Neolithic ............................................................................................................................ 59 Figure 6.5 CA pie chart plot of taxa, Neolithic ................................................................................................................. 59 Figure 6.6 Bar chart of cereal crop ubiquities, Neolithic................................................................................................... 59 Figure 6.7 Bar chart of legume ubiquities, Neolithic ........................................................................................................ 59 Figure 6.8 CA bi-plot of arable weed taxa, Neolithic ........................................................................................................ 60 Figure 6.9 CA pie chart plot of arable weed taxa, Neolithic ............................................................................................. 60 Figure 6.10 CA bi-plot of taxa Chalcolithic and early and middle Bronze Age................................................................ 61 Figure 6.11 CA bi-plot of taxa Chalcolithic and early and middle Bronze Age................................................................ 61 Figure 6.12 Bar charts of the ubiquities of the wheat crops .............................................................................................. 62 Figure 6.13 Bar charts of the ubiquities of the barley crops .............................................................................................. 62 iv

Figure 6.14 CA bi-plot of arable weed taxa, Chalcolithic and Bronze Age ...................................................................... 62 Figure 6.15 CA pie chart plot of arable weed taxa, Chalcolithic and Bronze Age ............................................................ 63 Figure 6.16 CA bi-plot of arable weed taxa, Chalcolithic ................................................................................................. 63 Figure 6.17 CA bi-plot of arable weed taxa, early and middle Bronze Age ...................................................................... 64 Figure 6.18 CA samples plot of arable weed taxa ............................................................................................................. 64 Figure 6.19 CA pie chart plot of arable weed taxa ............................................................................................................ 64 Figure 6.20 CA species plot of arable weed taxa .............................................................................................................. 65 Figure 6.21 Bar chart that shows the number of weed taxa over time ............................................................................... 65 Figure 6.22 Bar chart that shows flowering times, Cyprus ............................................................................................... 68 Figure 6.23 Bar chart of growing heights, Cyprus ............................................................................................................ 68 CHAPTER 7 Figure 7.1 Timeline of notable Cypriot cultural developments ......................................................................................... 77

v

LIST OF TABLES CHAPTER 1 Table 1.1 Chaîne opératoire of food and possible archaeobotanical datasets ..................................................................... 8 CHAPTER 2 Table 2.1 Chronology for Cypriot cultural phases discussed in text ................................................................................. 13 Table 2.2 Summary of differences between early Cypriot cultural phases ....................................................................... 22 Table 2.3 Summary of differences between Ceramic, Chalcolithic and Bronze Age cultural phases ............................... 23 CHAPTER 4 Table 4.1 Cypriot Archaeobotanical Publications List ...................................................................................................... 45 Table 4.2 Number of samples, volumes, and preservation and recovery methods for sites with Archaeobotanical data, Cyprus ............................................................................................................................................................................... 46 Table 4.3 List of taxa for each cultural phase that are not recorded in previous periods .................................................. 47 CHAPTER 5 Table 5.1 Table showing the average number of items per litre by trench, Kissonerga-Skalia ......................................... 52 Table 5.2 Table showing the average number of items per context type, Kissonerga-Skalia ........................................... 52 Table 5.3 Table of the average number of identifiable items per litre for sites located in Cyprus and the mainland Levant ................................................................................................................................................................................ 55 CHAPTER 6 Table 6.1 Ubiquities of the cereal crops ............................................................................................................................ 57 Table 6.2 Table of the arable weeds .................................................................................................................................. 65 Table 6.3 List of the total number of genera classified by plant family ............................................................................ 67 Table 6.4 List of the proportions of five plant families ..................................................................................................... 67

vi

LIST OF APPENDICES Appendix 1 Krittou Marottou-‘Ais Yiorkis context information ....................................................................................... 78 Appendix 2 Prastion-Mesorotsos context information ...................................................................................................... 79 Appendix 3 Souskiou-Laona context information ............................................................................................................ 80 Appendix 4 Kissonerga-Skalia context information.......................................................................................................... 81 Appendix 5 Identification criteria for cereals and non-cereal taxa .................................................................................... 83 Appendix 6 Selection of Images of charred Plant Specimens ........................................................................................... 86 Appendix 7 List of taxa recovered from Neolithic, Chalcolithic, and Bronze Age Cyprus .............................................. 89 Appendix 8 Calibrated Radiocarbon Dates from Cypriot Sites ....................................................................................... 100 Appendix 9 Krittou Marottou-‘Ais Yiorkis botanical data ............................................................................................... 105 Appendix 10 Prastion-Mesorotsos botanical data ........................................................................................................... 109 Appendix 11 Souskiou-Laona botanical data .................................................................................................................. 112 Appendix 12 Kissonerga-Skalia botanical data ............................................................................................................... 116 Appendix 13 Flowering Times and Growing Heights of Arable Weeds ......................................................................... 121 Appendix 14 Sites Names, Locations, Periods, and Phase Codes ................................................................................... 126 Appendix 15 Taxon Codes .............................................................................................................................................. 131

vii

ACKNOWLEDGEMENTS I would like to thank the University College London Graduate College and the Institute of Archaeology, University College London for partially funding the fieldwork in Cyprus. This book was completed under the European Research Council grant (ComPAg, no. 323842). I would like to thank my PhD supervisors, Professor Dorian Fuller, Dr Sue Colledge and Professor Cyprian Broodbank for their time and encouragement over the last few years. Thanks are due to Professor Dorian Fuller for his insight and comments and to Dr Sue Colledge for introducing me to flotation in the summer of 2005 and for all of her time, support, feedback, and genuine encouragement along the way. In Cyprus, the Field Directors, Alan Simmons, Eddie Peltenburg, Lindy Crewe, and Andrew McCarthy for allowing me to be a part of their excavations. Thanks are due to Paul Croft for his advice and flotation help. I also am grateful for my colleagues for great conversation and help over the years including Dr. Michelle Gamble, Dr. Lindy Crewe, Dr. Lisa Graham, and Dr. Andrew McCarthy. At UCL, I would like to thank my fellow archaeobotanists in room 306, Dr. Alison Weisskopf, Dr. Charlene Murphy, Dr. Chris Stevens, Ellie Kingwell-Banham, Rebecca Beardmore, Hanna Sosnowska, and Dr. Cristina Castillo. My husband, Thomas Lucas, for his support, good humour, and, most of all, patience; my mother, Vicki Switzer, for her never-ending support, for being my biggest fan and for always saying the right things to keep me going; my Dad, Andrew Espinda; my Grandma, Norma, and Grandpa, Earl; my sister, Cari, and her husband, James; and my in-laws, Anita and Doris, for making their home, my home away from home.

viii

CHAPTER 1 THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION 1.1 INTRODUCTION

1.2 RESEARCH AIMS AND OBJECTIVES

In the last thirty years there has been a great change in the understanding of the prehistory of Cyprus. Specifically, Cyprus has gone from an island on the periphery to cultural developments in prehistoric Southwest Asia to one that is now known to have played a key role in the spread of early agriculture. Prior to the late 1980s, the earliest evidence for human occupation on the island dated to c. 6,000 cal. BC. Since then evidence has demonstrated human activities on the island starting in the early 10th millennium BC (Simmons 2004). Current evidence from early Aceramic Neolithic sites on the island, dated to c. 8500 cal. BC, have put the island in the forefront of debates on the development and spread of agriculture in Southwest Asia. Recent data indicate that domestic cereal crops as well as animals, including evidence for the first commensals outside of the Levant and domesticated cat and dog, appear at nearly the same time on Cyprus as the mainland. The island is the first region to be colonized after the emergence of agriculture in the Levant and it was the earliest target for migration by farmers (Colledge et al. 2004; Lucas et al. 2012; Peltenburg et al. 2000; Peltenburg et al. 2003; Vigne et al. 2011a; Willcox 2003). Accordingly, Cyprus is a key region for research on the transition to, and spread of, agriculture into new areas. However, what happened after the introduction of agriculture to Cyprus has been relatively under researched. Further consideration is needed of the role of new crop introductions, local agricultural developments, agricultural intensification in subsequent cultural phases (i.e. the Pottery Neolithic, the Chalcolithic and the Bronze Age), and changes in food preparation and consumption technologies, which will provide evidence for interaction with surrounding geographical regions and increased social complexity. This research explores the dynamics of the spread of Near Eastern crop-based agriculture to the island of Cyprus, the development of the Cypriot prehistoric economy, including agricultural developments and cultural changes in food preparation and consumption technologies, and the influence of contemporary mainland Levantine cultural developments on these processes.

The primary objective of this research is the comparative quantitative analysis of the Cypriot archaeobotanical (i.e. plant remains recovered from archaeological contexts) record. This analysis includes data from four recently excavated sites on the island and a chronological and regional comparison of published evidence from Cyprus and the Near Eastern mainland; including archaeobotanical data from sites located in Iran, Iraq, Turkey, Syria, Israel, Jordan, and Egypt, and dated to the Pre-Pottery Neolithic (hereafter PPN or Aceramic Neolithic), the Pottery Neolithic (hereafter PN or Ceramic Neolithic), the Chalcolithic, and the Early and Middle Bronze Age. Archaeobotanical data from Krittou Marottou-‘Ais Yiorkis, Kissonerga-Skalia, SouskiouLaona, and Prastion-Mesorotsos, all located in Cyprus, will be used to address questions regarding the subsistence practices at the site-level and on a regional scale for the corresponding cultural phases, PPN, Early/Middle Bronze Age, Chalcolithic, and Late Ceramic Neolithic/Chalcolithic, respectively. Archaeological and artefact data from sites located in Cyprus and dated to the previously mentioned cultural phases will be used to address questions regarding changes in food preparation and consumption technologies. Data from Cyprus and comparative data from sites located on the mainland Levant will be used to address questions with respect to regional and chronological changes in agriculture and food culture and technology over time. 1.3 GENERAL REVIEW OF AGRICULTURAL ORIGINS AND ITS SPREAD TO CYPRUS Gordon Childe (1936) described the transition to agriculture, the “Neolithic Revolution” and the emergence of the urban state, the “Urban Revolution”, as the two most significant developments in human history. The former provided the likely foundation for surplus production which supported the emergence of the latter (Fuller et al. 2010, 13-14). It is remarkable that the transition to agriculture took place independently in multiple regions in the world beginning at roughly the same time (Price and Bar-Yosef 2011, 11). There are up to 24 (Fuller 2010; Purugganan and Fuller 2009) geographical regions in the world where agriculture developed independently and each region saw the domestication of a different suite (or ‘package’) of crops and animals and involved different chronologies, environments, technology, and cultural trajectories (Diamond 2002; Zeder 2006).

This chapter will present the research aims and objectives of this book followed by an introduction to the research questions addressed. An overview of the origins and development of crop-based agriculture will be provided followed by an introduction to previous interpretations of early Cypriot agricultural systems and food culture and food preparation and consumption technologies. The research questions and methods will then be presented and a chapter outline of the book provided.

Why the transition to agriculture began when and where it did has been extensively researched and explanations have been summarised by several authors, including Price and Bar-Yosef (2011), Harris (1996), Bellwood 1

 

CROPS, CULTURE, AND CONTACT

  (2005), Barker (2006), and Thorpe (1996). The transition to agriculture can be viewed in terms of domestication origins and dispersals of plants. The centers of origins (Vavilov 1926) or ‘hearths of domestication’ (Sauer 1952) are the regions where the plants and animals were domesticated and the secondary locations are the regions where agriculture and the ‘Neolithic package’ dispersed. This ‘package’ is characterized by regional variations in both plant and animal composition and exploitation patterns at the onset of the domestication process as well as in its dispersal (Conolly et al. 2011). In other words, the crops and animals did not spread as a complete ‘package’ as such, but rather each species has its own domestication and ‘dispersal story’ (Conolly et al. 2011; Colledge et al. 2004; Vigne; 2008; Zeder 2008). On the basis of previous interpretations the transition to agriculture in the Near East was considered to be a rapid process that involved a package of eight ‘founder crops’ which were domesticated in a ‘core’ region with a subsequent dispersal of this ‘package’ from the regional centre (Lev-Yadun et al. 2000; Zohary 1996, 1999; Abbo et al. 2010). However, in light of recent collated archaeobotanical data, a quite different picture emerges. Results of recent research are somewhat contradictory. They indicate that the process of plant domestication was slow (occurring over 3000 years), involved multiple domestication events and occurred in a more geographically dispersed area extending far beyond the previously held domestication ‘core’ area (Fuller et al. 2011).

and exchange between the two groups (Alexander 1978; Barker 2006; Bellwood 2005; Zvelebil 1996). On Cyprus the dispersal story of agriculture is one of demic diffusion, more specifically a targeted migration by farming populations from Southwest Asia in the Early PPNB (Colledge et al. 2004; Guiliane and Briois 2001; Peltenburg et al. 2000; 2003; Willcox 2003). As mentioned briefly above, less than thirty years ago it was thought that the population of the island by farmers occurred at a relatively late date, c. 6,000 cal. BC. Current research however, demonstrates a human presence before this time of hunter-gatherer populations during both the Epipalaeolithic (Akrotiri Phase) (Simmons 2004) and the Cypro-Pre-Pottery Neolithic A (McCartney et al. 2007, 2008; Vigne et al. 2011b) as well as agricultural communities during the Cypro-Early Pre-Pottery Neolithic B (Guiliane et al. 2011; Guiliane and Briois 2001; Peltenburg et al. 2003; Vigne et al.. 2011). The dynamics of the spread of farming to Cyprus demands investigation in light of recent evidence for continued occupation on the island, albeit ephemeral in the earliest phases. At this time what is known is that farmers from the mainland migrated to the island during the Cypro-Early PPNB. However, the extent to which possible residual hunter-gatherer groups interacted or assimilated with the PPNB farmers is not known. For example, there are indications that Cyprus continued hunting the Mesopotamian fallow deer alongside farming during the PPNB, PN, Chalcolithic, and Bronze Age which encourages an exploration of the dynamics of hunting and farming lifestyles.

Explanations for why hunter-gatherers became farmers have generally been categorised into either ‘push’ (i.e. food stress) or ‘pull’ (i.e. food choice with social motivations) models (Fuller 2003; Price and Bar-Yosef 2011; Barker 2006). That is, hunter-gatherers were either ‘pushed’ or ‘pulled’ into agriculture as a result of environmental stress (Childe 1936; Wagner 1977), population pressure (Boserup 1965; Flannery 1969; Binford 1968), or “pulled” through changing social structures (Hayden 1990, 1995; Cauvin 2000; Hodder 1990). Further, the dynamics of its dispersal have been argued to be a result of a mixture of either cultural (Edmonson 1961) and ‘demic’ diffusion (e.g. ‘wave of advance’) (Ammerman and Cavalli-Sforza 1971) or a mixture of the two (Alexander 1978; Colledge et al. 2004; Price 2000). The former involved the dispersal of domesticated crops, animals and technology with the adoption of farming by native hunter-gatherer populations (Zvelebil and Rowley-Conwy 1986) and the latter involved the spread of agricultural populations into new areas (Ammerman and Cavalli-Sforza 1971). Alexander (1978) discussed mobile (or ‘moving’) frontiers and stationary (or ‘static’) frontiers, which Bellwood (2000) later describes as ‘spread’ and ‘friction zones’, respectively (Alexander 1978; Bellwood 2000; Zvelebil 1996); the former developed during periods of agricultural expansions (e.g. colonisation by demic diffusion) and the latter developed in a more stable or gradually changing circumstances allowing for contact

1.4 AGRICULTURAL SYSTEMS AND CHANGE The creation of cultural artefacts involves a sequence of events to produce the end product, whether it is a stone tool, a ceramic vessel, or bread. Each stage of this chaîne opératoire produces waste, and thus evidence, for each stage of the sequence in the archaeological record. The nature of subsistence economies, which is agricultural systems, food processing and consumption technologies, can be thought of in terms of this chaîne opératoire. With regards to cereal agriculture, this sequence includes primary production, procurement, early processing, distribution, preparation (late processing), consumption, and disposal; each with its associated material culture and potential charred plant by-products (Table 1.1) (Fuller 2005; Goody 1982; Samuel 1999; Wilkinson and Stevens 2008). This section discusses the systems involved in the primary production, which are the agricultural regimes that occur pre-harvest. The different types of agricultural systems will be presented, as well as the dynamics of changes in these systems over time, including the primary motivations that drive agricultural change (i.e. intensification or extensification) and introductions of both new crops (e.g. through diversification) and agricultural technology (e.g. plough). Before a discussion on agriculture and agricultural systems can begin it is important to clarify the 2

 

CHAPTER 1 THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION terminology used in this book. Definitions of key terms, including cultivation, agriculture, domestication, intensification and extensification have been presented by many authors over the years (Fuller 2007; Fuller et al. 2011; Harris 1989; MacNeish 1992; Price and Bar-Yosef 2011) and the following definitions are a result of a combination of these. Hunting and gathering is the exploitation of wild plants and animals in their natural habitat and any modifications to either activity involves a very low investment of labour. Cultivation of wild plants involves sowing, harvesting, and re-planting in tilled soil. Cultivation has been identified in the archaeobotanical record by the presence of arable weed assemblages in association with wild cereals. The presence of arable weeds and wild cereals has been used to suggest what has been termed ‘pre-domestication cultivation’ in the Levant during the pre-pottery Neolithic (Colledge 1998, 2002; Hillman et al. 2001; White and Makarewicz 2011; Willcox 2011). The distinction between a wild and domesticated plant or animal is that the domesticated plant or animal differs genetically and morphologically from its wild progenitor species and is a result of either conscious or unconscious human intervention, specifically due to cultivation of plants and herding (i.e. the control of animal herds) or management of animals. Thus, as stated by Fuller et al. (2010, 2011) a primary difference between cultivation and domestication is that cultivation is something that people do, such as preparing the soil, sowing, and harvesting, while domestication is a genetic or morphological property of the plant that increases its adaptation to cultivation. However, other nonmorphological changes have been used to infer domestication status; including age profiles, milking, bone pathologies, and evidence for the spread beyond the natural habitats of the wild species (Meadow 1989; Price and Bar-Yosef 2011). Agriculture, therefore, involves the use of domesticated plants and animals for food and other resources.

cultivation, floodplain cultivation and intensive garden cultivation, all of which can be differentiated by permanence (e.g. fixed-plot or shifting cultivation, whether intensive or extensive), seasonality (i.e. spring versus autumn sowing of crops), and intensity (i.e. extensive versus intensive). With regards to seasonality, implications include both the amount of labour and time invested in cultivation practices, particularly with regards to the crop and livestock integration (i.e. manuring, etc…) and seasonal scheduling (autumn or spring sowing as in intensive garden cultivation). A method for determining these different agricultural regimes in archaeological contexts has been to use archaeobotanical assemblages of arable weeds in association with cereals, specifically using Functional Interpretation of Botanical Surveys (hereafter FIBS). FIBS provides a way of relating characteristics of plant species, particularly arable weeds associated with cultivation, to ecological variables to infer crop husbandry systems in the past and has been applied to both Near Eastern and European contexts (Bogaard 2004, 2005; Bogaard et al. 1999, 2001; Charles et al. 1997, 2002, 2003; Jones 2002; Jones et al. 1999, 2000; Charles and Hoppe 2003; Kreuz 2011). Shifting cultivation or slash-and-burn cultivation involves the burning of plots for soil regeneration and as a result the need for tillage and weeding is reduced. Shifting cultivation is considered to be extensive because of the low labour input per area (Bogaard 2004; Grigg 1974; Steensberg 1976). The reverse of shifting cultivation is fixed-plot cultivation, with cultural implications including claims to land and social inequalities (Bogaard 2004). Based on modern experiments aimed at determining the rate at which plant domestication occurred Hillman and Davies (1990) assumed that shifting plots was practiced, which is a more intensive form of shifting cultivation (Fuller et al. 2011). Another method is extensive ard cultivation (or the animal-drawn plough), which involves less human labour per unit area with lower yields. However, because a greater amount of land is cultivatable, the ability to produce surplus on a larger scale is increased (Bogaard 2004, see also Goody 1976; Halstead 1995). Hand cultivation with tillage, weeding and manuring is more intensive than both shifting cultivation and extensive ard cultivation because it involves high amounts of human labour per unit area. Intensive garden cultivation utilizes hand cultivation techniques and involves intensification methods including dibbling or row-sowing, handweeding or hoeing of crops, manuring and water management (Bogaard 2004 also refer to Halstead 1987, 1989; Jones 1992; Jones et al. 1999). Intensive mixed farming involves the integration of small-scale intensive garden cultivation and intensive livestock herding, with mutually exclusive benefits for both livestock and crops. The benefits for crops are that the animals provide manure for soil fertilization (whether it is collected manually or as a result of grazing) and the livestock help with tillage and the prevention of lodging. The primary benefit for the livestock is that the crop by-products can be used as fodder (Bogaard 2005).

There are differences in agricultural systems which have been discussed previously in terms of ‘intensive’ and ‘extensive’ methods. Pre-harvest ‘intensive’ methods refer to agricultural systems that involve high inputs of labour per unit area, resulting in high area yields. ‘Extensive’ methods refer to systems that involve smaller inputs of labour per unit of land exploited, resulting in smaller area yields (Bogaard 2004). Intensification and extensification practices aim to protect crops, maximise yields, and create some level of surplus, whether smallor large-scale. Intensification has generally been attributed to population increase, land limitations, or a combination of both these factors. Extensification, on the other hand, has been attributed to population increase in circumstances where land is not necessarily limited and expansion is possible (Wilkinson and Stevens 2008). With regard to past agricultural regimes in Neolithic Europe, Bogaard (2004) discusses four agricultural regimes, each with different economic and social implications. These are shifting cultivation, extensive ard 3

 

CROPS, CULTURE, AND CONTACT

  Models of agricultural regimes previously proposed for the Near East include Sherratt’s (1980) progressive system that begins with fixed-plot horticulture and progresses to flood-water farming and then to ploughbased agriculture with greater woodland clearance and large-scale irrigation systems. Recent research (Bogaard 2004, 2005) suggests that the most likely cultivation regime for early Neolithic sites in the Near East and Neolithic Linearbandkeramik (hereafter LBK) Europe is an intensive mixed farming system, which was previously proposed by Halstead (1987). This regime involved autumn-sowing of fixed-plots with high inputs of labour through tillage and weeding and integration between crops and animal husbandry through manuring of crops and animal grazing of fallow fields (Bogaard 2004, 2005). According to Bogaard (2004), Cyprus contributes to discussions on intensive mixed farming regimes, with evidence for this practice on the island by the end of the ninth millennium BC (Peltenburg et al. 2001). Beyond this statement, there has been very little research on early Cypriot agricultural practices and intensification to date.

the replacement of domestic sheep by a new domestic type. This research will explore the nature and impact of external contact by examining the possible various importation events along with evidence of non-native wild herbaceous taxa and domestic crops. An analysis of the non-native wild and domestic taxa has the potential to suggest origins and possibly define the direction from which farming spread to Cyprus (Colledge and Conolly 2007). 1.5 RELATIONSHIP OF EARLY AGRICULTURE TO FOOD CULTURE AND TECHNOLOGIES This section discusses food culture and technologies, particularly with regards to the procurement and the post-harvest processing of crops, including distribution, preparation, and consumption. The procurement of crops, including the material culture and plant by-products associated with the post-harvest processing activities will be introduced. A discussion of the prehistoric Near Eastern food culture and the technologies and material culture associated with distribution, preparation, and consumption will follow.

Intensification of crop-based agriculture on the island has been recognised in the Cypriot Chalcolithic to coincide with a general decrease in hunting and an increase in herding (Murray 1998). The evidence in Cyprus for a distinctive insular adaptation involving the controlled hunting and a system of game management of fallow deer throughout the Aceramic Neolithic, Ceramic Neolithic and Chalcolithic occupations has been discussed in previous publications (Croft 1991; Peltenburg et al. 2000). This economic strategy has been viewed as ‘de-intensification’ (i.e. decrease in agricultural practices) of agriculture on the island since the hunting of deer increases in importance, with fallow deer constituting over 70% of the Ceramic Neolithic and Early Chalcolithic assemblages (Clarke et al. 2007). Clarke et al. (2007, 62) suggest that the evolutionary pressures, including demographic pressure and sedentism, that drove agricultural intensification on the mainland were absent on Cyprus and as a result, deintensification occurred. The dynamics of this agricultural regime in connection with the hunting culture of Cyprus will be explored in Chapter 7.

In reference to prehistoric post-harvest early processing of cereal crops there are relatively few ways in which the sequence can be done (Hillman 1973, 1981, 1984; Jones 1987; Wilkinson and Stevens 2008). As stated by Wilkinson and Stevens (2008, 195) at its most basic the activities of early processing involve those that break things apart and those that separate things out, with the objective a semi-cleaned or fully cleaned grain that can be stored and further processed into food during late processing preparation. Since the goal is a semi-cleaned or cleaned grain store then it can be expected that the proportion of weed seeds to grain will lessen through the chaîne opératoire so the final stages will comprise grains and few weed seeds. Further, the weed seed by-products from each phase will be determined by their size, weight and their ability to separate from grain (e.g. chaff) (Fuller and Stevens 2009; Hillman 1981,1984; Jones 1984, 1987). The primary steps of early processing include threshing, winnowing, and coarse- and fine-sieving, each with signature plant by-products and material culture as evidence for that stage in the sequence (Table 1.1).

In a discussion of crop introductions to Cyprus, Colledge and Conolly (2007, 61) present the likelihood of multiple importation events of domestic plant taxa to the island in the Neolithic with increases in the number of domestic crops from the Cypriot Early Pre-Pottery Neolithic Phase B to the ceramic Neolithic. They reference Horwitz et al. (2004) and highlight the similarities between the botanical and faunal evidence, with the latter suggesting at least five separate importation events of wild animal species at Parekklisha-Shillourokambos (Colledge and Conolly 2007; Horwitz et al. 2004); however, Vigne (2009, 2011) argues for a more complex situation entailing multiple waves of introductions of wild and domesticated animals and considers the possibility of indigenous domestication of goats after 7500 cal. BC and

The next stage in the sequence is distribution, which includes the storing of the semi-cleaned or cleaned grain along with any residual weed seeds left over from coarse and fine-sieving. Artefacts associated with storage and distribution includes ceramic storage vessels, basketry (rarely preserved), silos and bins. Late processing of cereal crops includes the preparation of grain into food (e.g. from grain to bread). The material culture associated with late processing may include basketry, cooking vessels, ground stone tools (e.g. mortars and querns), hearths, and ovens. In addition to charred remains of food (e.g. grain or bread) evidence for what was consumed and the technologies involved can be inferred from serving containers (e.g. made of ceramic and stone). 4

 

CHAPTER 1 THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION As stated by Fuller (2005, 761), it is recognised that cuisine is regionally and culturally distinctive and thus, there are expectations that Indian, Italian, Japanese or French foods contain different ingredients and are prepared and consumed in different ways. Thus, culinary traditions in food preparation and consumption are evident cross-culturally (Goody 1982; Haaland 2007; Rowlands and Fuller 2010) and these food systems go back before the origins of agriculture to hunting and gathering populations (Rowland and Fuller 2010). Differences between food selection and preparation and consumption technologies have recently been examined, with roasting and grinding technologies characteristic of western Eurasia (i.e. the Near East and Mediterranean); boiling of porridge and brewing of beer with pottery in Africa; boiling and steaming in eastern Eurasia (i.e. China and the Far East); and a mixture of food cultures characteristic of South Asia (Fuller and Rowlands 2011, 37; Goody 1982; Haaland 2007; Rowlands and Fuller 2010). However, it is the Near Eastern trajectory of grinding, roasting, and baking that is most pertinent to an understanding of the Cypriot food culture because these processes are characteristic of the mainland Levant, and from where the earliest migrant farmers originated. To unravel the Near Eastern food culture the archaeological record is used in addition to the archaeobotanical data since artefacts often provide information on food production (e.g. ground stone for grinding grain to flour or cooking pots for boiling) and consumption (e.g. pottery for serving).

rectangular houses and an emphasis on storage facilities noted (Haaland 2006). With regards to pottery, Fuller and Rowlands (2011) argue that ceramics and grinding can be considered functional alternate adaptations for post-harvest intensification. In the Near East where ceramic technology developed after agriculture and grinding technology, the appearance of pottery was not associated with food preparation (i.e. boiling), as in Africa, but rather with food storage and consumption (i.e. cooking, serving, and drinking) (Atalay and Hastorf 2006; Haaland 2006, 2007). Thus, the introduction of pottery in the Near East meant new ways of preparation and consumption, including new ways of elaborating cooking (possibly mixed meat and plant dishes such as casseroles and stews) and serving (e.g. cups and small bowls) (Rowlands and Fuller 2010; Moore 1995, 47-48). The archaeological evidence of Cyprus differs in many respects from the cultures of the contemporary mainland (discussed in Chapter 2). These differences are evident in domestic architecture, with the persistence of circular structures up to the EBA, the timing of the introduction of pottery and appearance of ovens, changes in agricultural practices, and the reliance on deer hunting. The evidence suggests that the island’s colonists may “have retained many aspects of their earlier mainland lives” (Peltenburg 2004, 77). If the evidence demonstrates that Cyprus deviated from the mainland trajectory with regards to architecture and technological innovations during the PPNB, does it also show a divergence in food culture as well? This research will consider the archaeobotanical and artefactual evidence of Cyprus in an examination of the island’s early food culture and technology.

In the Near East there is an emphasis on the hearth, the oven, and the house as opposed to the pot in African traditions (Haaland 2006). There is substantial evidence for grinding of wild cereals (and other species) into flour to be baked into bread or bread cakes as early as the Epipalaeolithic, at Ohalo II (Piperno et al. 2004). Since grinding tools precede any evidence for pottery or ovens in this region it has been inferred that the bread was baked in open hearths, as ‘ash-baked’ bread (Haaland 2006). The earliest evidence for ovens (i.e. stone-filled, cylindrical pits) in this region comes from the PPNA at Mureybet (Cauvin 2000). Ovens become more widespread in later PPN levels, from ca. 7000 BC (Fuller and Rowlands 2011; Rowlands and Fuller 2010; also see Maisels 1990). Evidence from PPNA Jerf el Ahmar (dated to c. 9000 cal. BC) in particular provides insight into early Near Eastern food processing from an area interpreted to be a “kitchen” (Willcox 2002). The kitchen or preparation area contained archaeobotanical and artefactual evidence, including saddle querns, flat polished stone plates, limestone basins, a limestone bowl, pounding stones, and a hearth suggesting the soaking of barley, grinding of einkorn wheat and the preparation of seed cakes made from a mustard species with high seed oil content (i.e. Brassica/Sinapis) (Willcox 2002). There is evidence of circular bread ovens, or tannurs, in the Middle PPNB (Akkermans and Schwartz 2003; Wright 2000) and a correlation between the appearance of ovens to changing domestic architecture, with a change from round houses to

1.6 AGRICULTURE AND CYPRIOT BRONZE AGE

As introduced above, the relationship between agriculture and the emergence of the urban state is that agriculture was a precursor to urbanization because it provided the production of surplus (through intensification) needed to support craft specialisation, social hierarchies, trade, and the emergence of social complexity (Childe 1936; Diamond 2002; Fuller et al. 2010; Purugganan and Fuller 2009). As stated by Renfrew (1982, 267), ‘no complex society can function unless the level of subsistence production is sufficient to feed a range of specialists, including the leaders and organisers, in addition to those engaged in food production.’ There are three significant elements of agricultural production that supported trade and have been connected with the emergence of social complexity; these are surplus production, labour mobilisation and the production of “cash crops” (Sherratt 1999; see also Fuller and Stevens 2009). Evidence for urbanization in Cyprus comes from the Late Bronze Age. It marks a period of great social and economic change on the island, with substantial population increase, settlement in new regions of the island, changes in pottery, restructuring of craft-specialisation from domestic to large-scale 5

 

THE EMERGENCE OF THE

CROPS, CULTURE, AND CONTACT

  production, increased social complexity, enhanced international trade networks, in the copper industry in particular, and the emergence of hierarchy and the urban state (Steel 2004, 149-150). This book examines the archaeobotanical evidence for increasing social complexity leading up to urbanization in the Cypriot Late Bronze Age.

1981, 1999) and thus, as stated by Sherratt (1981, 263), the SPR is not just a matter of subsistence and economics but it represents a threshold in social development. The “cash crops” are those items that are grown for the purpose of trade and require specialisation in processing as opposed to crops used for household consumption (i.e. crops for ‘subsistence’) (Sherratt 1999). The stereotypic “Mediterranean triad” or “polyculture”, which consists of cereals together with wine and olive oil production, is seen as the classic Bronze Age economy and has traditionally been viewed as a marker of the ‘emergence of civilisation” (Hamilakis 1996, 1999; Renfrew 1972). However, Hamilakis (1999) argues that there is a lack of evidence in support of the Mediterranean polyculture in the Early Bronze Age. Since these “cash crops” are both risky and labour intensive, the industry is powered by market economies and ran by ‘wealthy’ farmers (i.e. individuals with additional resources) that can afford the risk associated with these crops (Forbes 1993; Hamilakis 1999). The origin of the market economy in Cyprus dates to the Late Bronze Age and therefore it is unlikely that the development of “cash-crops” on the island preceded urbanization.

Surplus production and the way in which communities store and process surplus can provide insight into increasing social complexity. A recent study by Fuller and Stevens (2009) illustrates how archaeobotanical data can contribute to evidence for changes in labour mobilization, or shifts in the structure of labour with regards to crop-processing at a site or regional level and further identify the relationship between agricultural production and social complexity. The degrees of labour mobilisation between focused (i.e. no more than single nuclear family), semi-large scale (i.e. extended family) and large-scale (beyond extended family) mobilisation is evident in the archaeobotanical record by evidence from storage practices and the timing of late processing of stored crops. The authors note that the storing of crops as ‘clean grain’ will require more labour and an ability to organise the workforce around harvest. In opposition, if the crops are stored with limited processing, the labour likely falls at the household level (Fuller and Stevens 2009). In Cyprus there is little evidence for intensification of labour and the control of surplus production prior to the Late Chalcolithic (e.g. Period 4, Kissonerga-Mosphilia) (Peltenburg 1993). Evidence of intensification of labour and surplus control comes from archaeological evidence of increased potential volumes in centralized storage containers (e.g. evidence from pottery) and a standardization and increase in crop processing tools (i.e. querns, mortars, and polished blades) (Peltenburg et al.. 1998). A re-assessment of the evidence for, and relationship between, increasing social complexity and labour mobilisation in Cyprus will be explored in this book.

1.7 RESEARCH QUESTIONS The questions addressed in this book have been introduced in the previous sections of this chapter. In this section there is summary of the questions to be addressed in the subsequent chapters. This research includes the archaeobotanical results from four recently excavated Cypriot sites, Krittou Marottou‘Ais Yiorkis, Kissonerga-Skalia, Souskiou-Laona, and Prastion-Mesorotsos. The following questions will be addressed for each site: Are there differences between samples and/or context types? What plant species are present in the samples? This research is the first to assemble and analyse the Cypriot archaeobotanical record from sites dated from the earliest settlers to the Bronze Age. The following research questions will be addressed using this dataset as a re-assessment of the Cypriot botanical record on a regional level: Does the botanical data provide evidence for one or multiple introductions of cereal crops to the island? What does the botanical data contribute to this idea of the spread of early agricultural ‘packages’? If there is evidence for multiple crop introductions, does it provide an index of varietal diversification through the adoption of new crops in the Cypriot prehistoric economy? What can the botanical data reveal about early agricultural practices in Cyprus, specifically with regards to permanence, seasonality, and intensity? Is there evidence of agricultural intensification or extensification, both pre- and post-harvest, prior to the Late Chalcolithic? Further, what does the botanical evidence contribute to debates of economic ‘de-intensification’ in light of the island’s unique hunting economy? With regards to food culture and technologies, artefactual evidence from excavations in Cyprus will be used to explore cultural

The secondary products revolution (hereafter SPR), first described by Sherratt (1981), involves a transition between a subsistence based on hoe-cultivation and animals kept for meat consumption to plough agriculture and the exploitation of animals for their secondary purposes. This revolution involves animals and their secondary products and services (i.e. milk, wool, transport abilities, and traction) and has recently has been argued to coincide with a tree crop and vine revolution, with perennial orchard crops and their secondary products (i.e. olive oil for consumption and ointments, wine, dried fruit, and textile fibers) (Fall et al. 2002; Fuller 2008; Sherratt 1999). As discussed by Greenfield (2010, 29-30) the innovations of this shift “led to a revolution in food production, mobility, local and interregional exchange and settlement patterns.” It is the surplus of these secondary commodities, the “cash crops”, that facilitated specialized markets with goods sold for profit and regional exchange in the emergence of complex societies (Fall et al. 2002; Fuller 2008; Sherratt 6

 

CHAPTER 1 THE ORIGINS, SPREAD, AND DEVELOPMENT OF AGRICULTURE IN CYPRUS: AN INTRODUCTION aspects of food preparation and consumption and changes in these over time. The following research question will be addressed: Does the artefactual and botanical data of Cyprus provide evidence for the Near Eastern grinding/roasting/baking food culture? Archaeobotanical data from Cyprus and comparative data from sites located on the mainland Levant will be used to address questions with regards to regional and chronological change in agriculture over time. The following questions will be addressed: Are there regional and/or chronological patterns evident in the archaeobotanical assemblages (i.e. crops and arable weed assemblages) of Cyprus and the mainland Levant over time? If so, what can these patterns reveal about interaction between the two regions during the AN, CN, Chalcolithic, and Bronze Age? Further, what can these patterns reveal about the emergence of social complexity in Cyprus and the development of the Cypriot Late Bronze Age economy? 1.8 ARCHAEOBOTANICAL SAMPLES A total of 8,721 litres from 217 samples were processed for the analyses of charred plant materials recovered from Krittou Marottou-‘Ais Yiorkis, PrastionMesorotsos, Souskiou-Laona, and Kissonerga-Skalia. All previously published archaeobotanical material from Cyprus, data from contemporary mainland sites, and the new botanical data presented here have been entered into an ACCESS database which will be used in the comparative analysis, which will be discussed in Chapter 3. 1.9 ORGANISATION OF THE BOOK This book is divided into seven chapters. Chapters 2 through 4 provide the environmental, archaeological, and archaeobotanical background for this research. Chapter 2 is an introduction of the environmental and archaeological context of Cyprus and the mainland Levant during the Aceramic Neolithic to the Middle Bronze Age. In Chapter 3 there is a discussion of the archaeobotanical methods used in the analysis of the archaeobotanical material from four recently excavated sites, the Cypriot botanical record, and in the regional comparison. Chapter 4 presents a summary and critical review of the archaeobotanical record of Cyprus. The results, a synthesis, and conclusions are provided in Chapters 5 through 7. Chapters 5 and 6 present the results and an interpretation, the conclusion, and suggestions for future research are presented in Chapter 7.

7

 

CROPS, CULTURE, AND CONTACT

   

Activity 

Material Culture 

Plant Material 

Pre‐harvest  Primary Production 

tilling, manuring, sowing 

plough   

 

Procurement     

harvesting  (by uprooting, plucking, or  cutting with sickle) 

chipped stone   (e.g. lithic blades, sickle  blades) 

chaff, grains, weeds (size depends on uprooting or  harvesting by sickle either high or low on culm) 

Post‐Harvest  Processing  (early processing)   

Glume  wheat  threshing ,  raking,  1st and 2nd  winnowing,  parching,  pounding,  coarse and  fine‐sieving 

chipped stone,  wooden poles,  basketry (sieves),  mortars  

Distribution  

storing and allocation 

storage vessels, silos, pits,  bins 

Free‐threshing wheat  Glume wheat   threshing →  threshing → grain and light  spikelets, awn fragments  chaff fall free, rachis  raking →  segments  spikelets, awns, culm  raking→   bases, weed seeds  free‐grain, fine chaff,  winnowing→ grain, heavy  rachises, weed seeds  straw nodes, rachis  (waste straw for fuel  fragments, spikelets, most  fodder and temper)  weed seeds (light and  winnowing→  small) (residuals used for  grain, heavy straw nodes,  fuel, fodder, temper)  rachis fragments, spikelets,  Coarse‐sieving→ prime  most weed seeds (light and  grain, heavy weeds,  small) (residuals used for  spikelet forks  fuel, fodder, temper)  fine‐sieving→  Coarse‐sieving → heavy  prime grain, some weeds  weeds, grains, spikelets  similar in size, and rare  fine‐sieving → prime grain,  some weeds similar in size,  rachis fragments (tail grain  and small weed seeds and  and rare rachis fragments  heavy chaff used for animal  feed and famine foods)  Cleaned or semi‐cleaned grains including weed seeds 

Preparation   (late processing)   

grinding  baking 

grains   similarly‐sized weed seeds, with a predominance of larger  weeds over small weed seeds 

Consumption   

eating/drinking 

hearths, ovens  ground stone  (e.g.  mortars, querns), cooking  vessels  pottery,  stone bowls, and platters  

Disposal  

clearing up 

preparation waste 

 

Free‐threshing  wheat  threshing ,   raking,  winnowing,   coarse and fine‐ sieving 

rarely preserved food remains  

Table 1.1 Chaîne opératoire of food and possible archaeobotanical datasets for cereals (taken from Goody 1982, 37; Fuller 2005, 762; Samuel 1999, 124, 130; Hillman 1981, 1984; Jones 1987; Wilkinson and Stevens 2008; Fuller et al 2010; Fuller and Stevens 2009)

8

 

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT There are three Chalk Plateaus located in the south of Cyprus, named the Paphos, Limassol, and Lefkara Plateaus. The Polis Lowlands is within the Chalk Plateaus but Christodoulou (1959) distinguishes based on morphology, which includes raised beaches, sand-dunes in the eastern portion, a gorge, and numerous river terraces. The Ktima Lowlands measure ca. 11 miles wide and runs ca. 26 km along the southwest coast from the villages of Kissonerga to Kouklia. The Limassol Lowlands include the Akrotiri peninsula in the southern portion, the delta of the Kouris River in the north-west, and living sand-dunes in the northeast coast.

2.1 INTRODUCTION This chapter includes a discussion of both the environmental and archaeological context of this research. First, an examination of the study area will be presented including a short discussion of the topography, geology, climate, and vegetation of Cyprus and a more detailed summary of the current and past vegetation of the areas from which new archaeobotanical data is analysed in this book. Given that this research looks at chronological and regional change in the economy of prehistoric Cyprus it is important to situate the island within its broader regional context. The second section of this chapter outlines the archaeological background of this research and includes a summary of the archaeological complexes of Cyprus and the Levantine mainland dated to the Aceramic Neolithic, Ceramic Neolithic, Chalcolithic, and Early and Middle Bronze Age. However, the primary focus of this research is Cyprus and accordingly the cultural background of contemporary mainland traditions will be discussed only in more general terms. 2.2 CYPRUS, THE STUDY AREA TOPOGRAPHY AND GEOLOGY Cyprus is situated in the eastern Mediterranean about 105 km west of Syria and 65 km south of Turkey. It is located between 34°33’ - 35°41’ N/32°17’ - 34°35’ E. The island is the third largest in the Mediterranean, third to Sicily and Sardinia, with an area of about 9251 square km (Christodoulou 1959). Christodoulou (1959) separated the island into nine topographically defined regions and more recently Zohary (1973) and Meikle (1977) collated them into four: 1) The Coastal Belt; 2) The Kyrenia or Northern Range; 3) The Troödos or Southern Range; and 4) The interior lowland Mesaoria or Central Plain (Meikle 1977, 1-3). This discussion is structured according to the four classifications outlined by Zohary (1973) and Meikle (1977) with the inclusion of the sub-regions discussed by Christodoulou (Figure 2.1).

Figure 2.1 Map of Cyprus showing general geology and topography and botanical divisions after Meikle 1977 (map modified after Zohary 1973)

The Northern or Kyrenia Range is an alpine mountain range that measures ca. 80 km long and runs parallel and near to the island’s northern coast (Meikle 1977; Christodoulou 1959). The land is mostly uncultivated and the only cultivated areas are located in the upper valleys. Christodoulou (1959) states that the peninsula is a continuation of the Kyrenia mountain range but the folding is much less severe and thus, the peninsula is separated from the mountain range based on morphology. He describes the region as “one of beveled ridges, flat-topped plateaus, wide basins or valleys, steep banks the result of undermined limestone capping” (Christodoulou 1959, 10).

The coastal belt consists of fertile, tilled land as well as uncultivated land and is divided into the following subregions: the Larnaca Region, the Polis Lowlands, the Ktima Lowlands, the Limassol Lowlands, and the Chalk Plateaus of the south, with the Larnaca region further divided into the upland and lowland areas. The upland area includes portions of the lower Troödos Massif and the pillow-lava hills and the lowland region consists of raised beaches and sand dune formations (Christodoulou 1959). The coastal belt is low-lying and is characterized by mostly rocky or stony shores with some sandy bays and salt-flats (Meikle 1977).

The Troödos mountain range is part of an old mountain system that is in alignment with the African Rift Valley system and is the most defining geological feature of Cyprus, encompassing the central igneous mass and covering much of the southwest of the island (Christodoulou 1959; Meikle 1977). Most of the mountain range lies above 1200 meters in elevation and the highest point of the island is Mt. Olympus at about 1950 meters elevation (Meikle 1977). The higher elevation comprises igneous rock of dolomite (i.e. igneous dolerites and gabbros), which is surrounded by 9

CROPS, CULTURE, AND CONTACT pillow lavas in the lower elevations; the pillow lavas surround the Massif but cluster primarily in the NE (Christodoulou 1959; Constantinou 1982). The pillow lavas are also found in small outcrops elsewhere on the island (Christodoulou 1959) and are the main source of two important resources exploited in Cypriot prehistory, namely copper and picrolite (Peltenburg 1982b; Steel 2004).

year but varies significantly according to elevation and topography, with the hottest summer temperatures occurring in the Central Lowlands and the coolest in the highest elevations in the Troödos mountains (Christodoulou 1959). VEGETATION Meikle (1977, 4-8) divides the island into eight phytogeographic zones (Divisions 1-8) which are covered by six general vegetation groups; pine forests (i.e. Pinus brutia and Pinus nigra), garigue (or the more dense, maquis), rocky areas, coastal areas, wetlands, and cultivated land (Figure 2.1) (Tsinides 1998, 11-20). More specifically, Zohary (1973, 151) describes the vegetation of Cyprus as a classic Mediterranean phytogeographical region and discusses five general vegetation classes: 1) the Quercetalia calliprini (i.e. Quercetalia and Sarcopoterietalia), 2) the Quercetea cerris orientalia, 3) Cedretea, 4) the central plain, and 5) the narrow alpine zone of the Troödos Mountains (Zohary 1973, 153). With the exceptions of the Quercetea calliprini and the central plain, all categories are located above 900 meters elevation and have, to date, little evidence for human occupation in prehistory. As a result this discussion will focus primarily on the Quercetea calliprini since it is this class that best describes the present vegetation of the location of the four sites with new botanical material analysed in this research, which are located in the coastal belt of the island’s southwest (Division 1) below 400 meters in elevation.

The Mesaoria or Central Plain runs about 90 km west to east and is 56 km wide in the east and 29 km in the west (Christodoulou 1959; Meikle 1977). It separates the Kyrenia and Troödos mountain ranges (Meikle 1977). Christodoulou (1959) argues that the region should be considered the Central Lowlands as opposed to the Central or Mesaoria Plain because it is technically not a plain and the region is not exclusively between the mountains as the term “Mesaoria” implies (i.e. two-thirds of the region lie east of the point at which the Troödos mountain range recedes) (Christodoulou 1959). This region is for the most part tree-less but fertile with cereal fields, traversed by seasonal rivers that are dry in the summer months (Meikle 1977). PRESENT CLIMATE Cyprus has an arid Mediterranean climate with short, cool and wet winters and long, dry and hot summers. The annual rainfall ranges from 250 mm to 1270 mm per year, with an island-wide average of approximately 500 mm (Christodoulou 1959; Meikle 1977). Regional variations in both temperature and rainfall are a result of differences in elevation, topography, and season (Christodoulou 1959; see also Meikle 1977, 1-3). Most of the rainfall occurs at higher elevations between October and March. At the topmost slopes of Mt. Olympus the average annual rainfall is approximately 1000 mm compared to less than 254 mm in areas of the western Mesaoria (Michaelides et al. 2009) (Figure 2.2).

Following Meikle (1977, 4), a large portion of the island was likely forested in antiquity, and with the exception of agricultural land, is the main vegetation cover of the island today. The only area that is not covered by pine forest is in the Mesaoria plain towards Morphou (Tsinides 1998). The Troödos mountain range (Division 2) is the most significant geographical feature of the island, with an elevation mostly above 1200 meters. Although forest cover has likely receded since the Neolithic, Chalcolithic, and Bronze Age occupations due to deforestation, there is still a large part of this region covered by golden oak and cedar (Quercus alnifolia and Cedrus libani ssp. brevifolia, respectively) and in the lower slopes, pine forest (Meikle 1977, 4). The latter of which was likely to have covered the prehistoric landscape around Krittou Marottou-‘Ais Yiorkis, Prastion-Mesorotsos, Souskiou-Laona, and KissonergaSkalia. In her report on the wood charcoal macroremains from Neolithic and Chalcolithic occupations at Kissonerga-Mosphilia, Asouti (1998, 74-76) characterised the prehistoric landscape as typical maquistype Mediterranean dense woodland/forest vegetation with a range of evergreen and deciduous oak taxa including oak (Quercus sp.), lentisk (Pistacia sp.), wild carob (Ceratonia sp.), fig (Ficus sp.), pine (Pinus sp.), and olive (Olea sp.). However, today modern-day villages and agricultural land are the dominant features surrounding these sites, with the presence of cultivated

In regards to the island’s seasonal rainfall Christodoulou (1959, 21) states the winter rainfall is probably the most significant aspect of the island’s environment. Winter receives the greatest amount of rain in the year and it is both beneficial and effective because it occurs when temperatures are lower, lessening the amount of moisture lost by evaporation. Accordingly, summer rainfall would be much less effective and even damaging, causing floods, soil erosion, crop damage, and higher malaria rates due to the fact that mosquitos could breed in areas of standing water (Christodoulou 1959). There is variability in annual rainfall year to year and prolonged droughts on the island are not infrequent and can last decades (Meikle 1977, 1-3). Christodoulou (1959) emphasizes the effect of drought on the island’s land-use patterns and people. Recurrent droughts have been influential in the stability of the island’s economy in the recent past and, no doubt, also in prehistory. During drought, crops may fail, springs and wells may dry up, livestock may diminish, and settlement relocation may result. Variability in temperature is minimal from year to 10

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT cereal fields, fallow fields, vineyards, and citrus and banana groves.

Potamogeton sp., Tamaricetum pentandrae, Alnetum orientalis, Platanetum orientalis, and Salicetum albae (Meikle 1977; Zohary 1973, 153-154).

The current vegetation of the areas surroundings these sites is primarily the evergreen maquis and forests of Quercetea calliprini (i.e. the Quercetalia and Sarcopoterietalia), which extends from sea level to about 1200 meters elevation and comprises maquis, batha, and garigue plant communities. This oak-type maquis vegetation of tall shrubs (4-6 meters) includes the species Phillyrea media (broad-leaved phillyrea), Styrax offinalis (styrax), Olea europaea (olive), Arbutus andrachne (Greek strawberry tree), Pistacia terebinthus (terebinth), Laurus nobilis (laurel), Quercus coccifer (kermes oak), and Rhamnus alaternus (Mediterranean buckthorn) (Meikle 1977, 4; Zohary 1973, 153-155). The Quercetalia class comprises Ceratonieto-Pistacietum lentisci, Quercus calliprinos-Pistacia palaestina maquis, and Pinetum brutiae (Zohary 1973, 153). The Ceratonieto-Pistacietum lentisci, or carob-lentisk community, covers the dry and lower zones of the island and extends in elevation to about 300 meters. As the name suggests, the carob tree (Ceratonia siliqua) and lentisk (Pistacia lentiscus) are the principal components of this vegetation.

PALEOENVIRONMENT OF THE NEAR EAST AND CYPRUS “…the conventional wisdom that the increases in social complexity associated with the development of agriculture, large settled communities and the earliest states were made possible by relatively stable, benign climatic conditions during the Holocene can no longer be upheld… It therefore appears increasingly likely that the astonishing increases in the level of human social complexity documented in the archaeological record since the beginnings of sedentism in the late Pleistocene were accompanied by profound and recurrent climatic and environmental challenges.” (Wasse 2007, 46-47 citing Brooks 2006, 26) With regards to the island’s early prehistory, it is important to situate the archaeological evidence for increasing social complexity in an environmental framework (following Wasse 2007). There are multiple lines of evidence used in the reconstruction of paleoenvironments, specifically proxy data that provide inferential evidence for climate change. In addition to bio-archaeological data (i.e. archaeobotany, zooarchaeology, human osteology), other sources can be used to infer climatic conditions, including glaciological/ice cores (i.e. oxygen and hydrogen isotopes), geological (i.e. marine core sediments, microfossils, oxygen isotopes), terrestrial (i.e. landforms and deposits), biological (i.e. tree-rings, pollen, plant macrofossils, vertebrate and invertebrate fossils), and historical climate records (Roberts 1998; Rosen 2007, 17-18; Wasse 2007). There are limited available data for early Holocene environmental reconstruction of Cyprus (Stanley-Price 1980, 1) and these data are restricted to extrapolations from the present environment of Cyprus and from paleoenvironmental research in adjacent countries on the mainland (Stanley-Price 1980; Wasse 2007). Rosen (2007, 17-31) provides a comprehensive summary of the common methods used in the reconstruction of the Near Eastern Late Pleistocene and Holocene climate, including historical records, pollen analysis, isotope analysis and geomorphology. Syntheses of these numerous lines of evidence have been provided by Baruch and Bottema (1999), Goring-Morris and Belfer-Cohen (1997), Hillman (1996), Robinson et al. (2006), Zohary (1973) and for Cyprus, Wasse (2007).

The deforested and heavily grazed maquis landscape has developed into the Sarcopoterietalia class of Quercetea calliprini, which comprises batha and garigue communities of Sarcopoterium spinosum (pricky burnet), Coridothymus capitatus (conehead thyme), Calycotome spinosa (spiny broom), Lavandula stochas (lavender), and Cistus spp. (rockrose) (Zohary 1973, 153). These areas are more common and are covered by the less dense, garigue communities of low (< 3 meters) shrubs including, rockrose, Genista sphacelata, Calycotome villosa, gromwell (Lithospermum spp.), and mastic tree (Pistacia lentiscus) (Meikle 1977, 4). Further, where animal grazing is even more intensive the garigue vegetation is reduced to batha, which includes the following species: thorny burnet, needle sunrose (Fumana arabica), Micromeria spp., Teucrium polium (felty germander), Rhamnus oleoides (mediterranean buckthorn), Salvia spp. (sage), Lavandula stochas, Genista fasellate, and thymus (Thymus capitatus) (Meikle 1977, 4; Zohary 1973, 154). Additionally, there are aquatic plant communities near rivers and springs, which are common throughout the coastal belt and near to the aforementioned sites. For example, there is a spring less than 300 meters south of Krittou Marottou‘Ais Yiorkis and the Ezousas River is located only 1 km from the site, and both Souskiou-Laona and PrastionMesorotsos are located in the Dhiarizos River Valley. It is near these river banks and springs that aquatic plant communities of the Phragmitetea and Potamotea classes persist today and also likely to have been present in the past. The Phragmitetea class is dominated by Phragmites and sedges (Cyperaceae) including Phragmites australis, Cyperus spp., Eleocharis palustris, and Carex spp. The Potamotea (pondweed class) can be found near river banks and is dominated by aquatic species including

The climate of both the Late Pleistocene and the Holocene is characterized by fluctuating glacial and interglacial episodes and periods of moist climate and severe drought and it is within the fluctuating environmental context that significant cultural changes occurred. There are three time periods from the Late Pleistocene to the Late Holocene where climate is argued to have had an impact on the cultural traditions of the mainland Levant, the Terminal Pleistocene (i.e. Younger Dryas, hereafter YD), the Mid-Holocene, and the Late 11

CROPS, CULTURE, AND CONTACT Holocene (Rosen 2007). However, recent research by Maher et al. (2011) discusses the lack of sound evidence in support of climatic events as the trigger for cultural change. They argue that the relationship between climatic instability and cultural change (i.e. changes in subsistence, social organization, and technology) is more complex than a cause and effect model (Maher et al. 2011). They do not discount that cultural change occurred in climatic instability only that is was not the direct cause. With regards to Cyprus, Wasse (2007, 48) states that there is little reason to accept that the island would not have experienced recurrent climatic fluctuations and that the Cypriot cultural tradition developed within this dynamic environmental context.

2.3 ARCHAEOLOGICAL COMPLEXES OF CYPRUS AND THE MAINLAND LEVANT CHRONOLOGY This research covers the duration of the Aceramic Neolithic, Ceramic Neolithic, Chalcolithic, Early and Middle Bronze Age (i.e. Early and Middle Cypriot) cultural entities of Cyprus. The Late Bronze Age (i.e. Late Cypriot) botanical material will be discussed briefly for comparison and to highlight the Middle Cypriot period as the end of small-scale, pre-urban Cyprus. Based on relative and absolute chronologies, the cultural entities of Cyprus have been subdivided into the Akrotiri, Aceramic Neolithic, Ceramic Neolithic, Chalcolithic and Bronze Age (Table 2.1, refer to Figure 2.3 for site locations). All radiocarbon dates presented here are calibrated unless otherwise discussed. The Aceramic Neolithic is further subdivided into the recently defined Cypro-Pre-Pottery Neolithic A (hereafter Cypro-PPNA), the Cypro-Early Pre-Pottery Neolithic B (hereafter Cypro-EPPNB), Cypro-Middle Pre-Pottery Neolithic B (hereafter Cypro-MPPNB), Cypro-Late Pre-Pottery Neolithic B (hereafter the Cypro-LPPNB) and the Khirokitian (Peltenburg 2003; Clarke et al. 2007). The Ceramic Neolithic follows a gap in archaeological evidence from about 5300 to 4750 BC and spans the period c. 4750 to 3900 BC (Clarke et al. 2007). Based on excavations at Chalcolithic Kissonerga-Mosphilia, Peltenburg et al. (1998) have subdivided the Chalcolithic into an Early (Kissonerga-Mosphilia Period 2), Middle (Kissonerga-Mosphilia Period 3) and Late (KissonergaMosphilia Period 4) occupation. The Bronze Age is subdivided into a Philia phase, an Early Cypriot I-III, and a Middle Cypriot I-III phase (Frankel and Webb 2006). Presented here is a summary of the major themes of the archaeological complexes along with a brief overview of the material culture including artefacts associated with the processing of food (i.e. pestles, mortars, and querns) and those that have been used to infer external contact and interaction with the mainland. The archaeological complexes of the mainland Levant are summarised under their chronologically contemporary Cypriot cultural traditions, “which bear only the smallest relation to chronological and cultural periodizations on the mainland” (Clarke 2007, 9). Further, as Clarke (2007, 9) highlights, there are inconsistencies in local and regional chronologies and cultural sequences across the eastern Mediterranean region and these have enforced researchers to study regions in isolation. In consideration of this, I have kept the summary of the mainland archaeological complexes brief, highlighting the main differences between the cultural traditions of the island and the mainland in an attempt to discuss significant events within the broader interaction sphere over the course of Cypriot cultural developments.

The Terminal or Late Pleistocene is characterised by a warming trend after the Last Glacial Maximum (hereafter, LGM), which includes a combination of higher temperatures and humid conditions of the BøllingAllerød (15-13.0 cal. k BP). Following this period is the YD (12.7-11.5 k cal. BP), which was a return to more cold and dry conditions that lasted about a millennium. Subsequent to the YD was a return to moister and warmer conditions at the Pleistocene - Holocene transition (Rosen 2007; Robinson et al. 2006; Moore et al. 2000). It is at the close of the YD that Levantine exploration to the island of Cyprus began, with the first evidence of human activity at Akrotiri (discussed below). The YD is considered to be a possible catalyst for the transition from a subsistence based on hunting and gathering of wild resources to the cultivation of domesticated plants and animals (Bar-Yosef 1998, Moore and Hillman 1992; also see Colledge and Conolly 2011). The subsequent Early Holocene (9500-5500 cal. BC) is a period of more moist and warm climatic conditions and it is during this time that a there were changes in human culture, including the origins and spread of the first agricultural villages (Rosen 2007). At ca. 6000 cal. BC, during the Mid-Holocene, there was an increase in dry periods and it is at this time that the first complex societies in the Near East developed. Rosen (2007, 7) states that even in these moist periods, periodic drought was a threat and restrictive element that affected cultural development. This is also the case for Cyprus in the most recent past and has been noted by Christodoulou (1959). He highlights the impact of recurrent drought on the island’s communities and states the instability of rainfall from year to year not only has an effect on the psyche of the Cypriot peoples but it has an effect on settlement patterns and distribution with relocation of communities a common result. The Late Holocene climatic amelioration at ca. 4000 cal. BC is characterized by a return to more favorable climatic conditions and from about this time period onward there was great population growth, the rise of empire states, an increase in agricultural intensification, and the cash crop market economy (Rosen 2007), all of which there is archaeological evidence for in Cyprus and discussed below.

12

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT and the exploitation of wild cereals (Bar-Yosef 2001; Kislev et al. 1992). AKROTIRI PHASE OF CYPRUS The Akrotiri Phase is the earliest phase of human activity discovered on Cyprus and is described as an exploratory period when the island was visited repeatedly by transitory hunter-gatherers (Peltenburg et al. 2001, 62; see also Broodbank 2006, 209) in the early tenth millennium BC (Simmons 2004). Akrotiri-Aetokremnos, or Vulture Cliff, lies on the southern coast of the island on the Royal Air Force Base (Akrotiri peninsula) and is the site after which the cultural phase is named. The site is thought to have been a short-term occupation and 31 radiocarbon determinations place the occupation around 9825 BC (Simmons 1999). The chipped stone industry, which comprises thumbnail scrapers, burins, retouched blades, bladelets and flakes, parallels Levantine Epipalaeolithic and early Neolithic assemblages (Simmons 2004), particularly the chipped stone from Öküzini in Antalya (Bar-Yosef 2001, 37). In addition to the pygmy hippopotamus, the faunal assemblage also includes dwarf elephants, deer, pig, genets, mice, birds and marine invertebrates. Other material culture includes stone and shell beads. Of note is that in contrast to subsequent cultural phases no obsidian has been reported on the island at this time (Simmons 2004).

Table 2.1 Approximate chronological phasing for periods in Cypriot prehistory covered in this research (Peltenburg et al. 2000; 2003; McCartney 2004; Clarke et al. 2007; Knapp 2008; Simmons 1999; Frankel and Webb 2006; Steel 2004; Manning et al. 2010) (** denotes climatic event discussed in text)

AKROTIRI PHASE OF CYPRUS AND THE NATUFIAN OF THE MAINLAND LEVANT

Additionally, stone tools and hearths have been found in association with faunal remains of the indigenous pygmy hippopotamus, Phanourios minutus (Simmons 1999). Simmons argues that humans played an active part in the extinction of the fauna (Simmons 2001) and for this, association between the archaeological and the faunal data are controversial. Others have questioned the links between the humans and the pygmy hippos, specifically whether humans were responsible for the killing of the hippos at the site (Ammerman and Noller 2005). Regardless of whether or not humans played a part in the extinction of this fauna at Akrotiri-Aetokremnos, the data from the site fueled discourse and further inquiry into an earlier phase of human occupation than the previously held Khirokitian (Ammerman and Noller 2005; Simmons and Mandel 2007). Ammerman and colleagues have undertaken surveys with the specific intention of demonstrating that Akrotiri-Aetokremnos is not the only site that represents the earliest phase of human activity on the island (Ammerman et al. 2006; 2008). Investigations of coastal sites located in the Akamas in the west and near Agia Napa in the east; these sites lie beneath hardened sand dunes and have provided evidence of a flake-based technology with the reduction of chert pebbles and small blades (Ammerman et al. 2008). The first radiocarbon dates from Nissi Beach place the site in the Cypro-MPPNB, which has been argued to be the result of inversion of cultural deposits caused by the sea. Despite evidence to the contrary, Ammerman et al. (2008) argue the lithic assemblages at Nissi Beach, Aspros and Alimman parallel mainland Natufian small flake and blade assemblages.

THE NATUFIAN OF THE MAINLAND LEVANT The earliest recorded island travellers were part of a mainland Levantine cultural complex known as the Natufian (ca. 12500-9500 cal. BC). The Natufian is divided into three phases, an Early, Late and Final Phase and is thought to be the link between hunter-gatherers and the first farmers (Rosen 2007) and is what Bar-Yosef (2011) has described as the “point of no return” (BarYosef and Belfer-Cohen 1989; Belfer-Cohen and BarYosef 2000), meaning it was the point at which huntergatherers continued on the path to a farming lifestyle. It is composed of semi-sedentary small villages, with evidence for storage based on the presence of commensal animals (i.e. house mouse) (Tchernov 1991; Bar-Yosef 2011). Evidence for human occupation includes circular pit-houses with stone foundations and external secondary burials, with evidence for decorated human skeletal remains (Bar-Yosef 2011). Material culture includes artefacts interpreted as symbolic representations, ground stone, including pounding tools (mortars, bedrock mortars, robust pestles) and milling and grinding tools associated with the processing of plant material (handstones, grinding slabs) (Wright 2000), microlithic chipped-stone assemblage, including cores, bladelets, flakes, sickle blades, with evidence of hide working from the chipped stone tools, basketry, and bone hooks (i.e. for fishing). The Natufian subsistence consisted of animal hunting (gazelle and small animals including hares, tortoise, gazelles, fallow deer, roe deer, and fish) 13

CROPS, CULTURE, AND CONTACT PRE-POTTERY NEOLITHIC A OF CYPRUS AND THE MAINLAND LEVANT

excavations of large sites, as opposed to small artifact scatters (McCartney et al. 2006). The project has surveyed eleven sites near the modern villages of Pera Chorio (i.e. Agia Varvara, Politiko, Analiondas, and Alampra). Excavations at Agia Varvara-Asprokremmos have provided evidence for PPNA occupation on the island filling the previous gap in occupation between the Akrotiri Phase and the Cypro-EPPNB. Subsequently, the evidence has embedded Cyprus in the mainland PPNA interaction sphere with parallels in material culture (Manning et al. 2010). The site includes evidence of refuse in natural hollows (i.e. pits) and a simple semisubterranean pit shelter with a posthole. The material culture recorded at the site includes picrolite pendants and beads, dentalium shell beads, ground stone artefacts, stone vessels (i.e. flat based ‘trays’, hemispherical bowls, one ochre-painted), two ground stone shaft-straighteners (parallels PPNA Jerf el Ahmar), and a fragment of a baked clay anthropomorphic figurine (Manning et al. 2010). Evidence from the chipped-stone assemblage includes microliths, bifacially backed blades, over 100 arrowheads, an absence of naviform core technology and a presence of unidirectional lithic core reduction similar to sites dated to the late PPNA/EPPNB in the northern Levant (McCartney et al. 2006; Manning et al. 2010). Similar to the preceding Akrotiri phase, there is evidence of wild pig management and the exploitation of fish and fresh water crab (Manning et al. 2010). Manning et al. (2010, 703) state that although the island did not participate in the earliest obsidian trade networks, evidence from the site is suggestive of an interaction extending from the Euphrates to the southern Levant (Manning et al. 2010, 703). Additionally, recent excavations at Ayios Tychonas-Klimonas provide evidence that further establishes the presence of a PPNA on Cyprus. Ayios Tychonas-Klimonas is located on the other side of the Troödos mountain range from the EENC sites and the architecture includes a communal 10m circular subterranean structure with multiple hearths and post-holes and several circular domestic structures (Vigne et al. 2011b; Vigne et al. 2012). Faunal remains so far demonstrate the exploitation of wild boar, the presence of domesticated cats and dogs, and evidence of commensals, suggesting possible storage. The chipped stone assemblage includes burins, scrapers, drills, glossed sickle blades, and stone shaft-straighteners. Evidence of PPNA sites with parallels in material culture and architecture to the mainland demonstrates connection and interaction between the two prior to the Cypro-PPNB.

PRE-POTTERY NEOLITHIC A OF THE MAINLAND LEVANT The PPNA on the mainland has the first undisputed evidence for village settlements and the cultivation of cereals (Colledge 1998, Hillman and Davis 1990). BarYosef (2001) describes the PPNA as a non-egalitarian agricultural society with hunting and gathering. Sites dated to the PPNA in the Jordan Valley include Jericho, Netiv Hagdud, Iraq ed-Dubb, Wadi Feinan and Nahel Oren, and Jerf el Ahmar and Mureybet II in the Middle Euphrates (Bar-Yosef 2001). The architecture of the mainland PPNA is comparable throughout the region, comprising circular unbaked mud-brick pit-houses with stone foundations typically measuring about 10-20 feet in diameter (Bar-Yosef 2001; Rollefson 2003). The ground stone assemblage includes shallow grinding bowls, hand stones, mortars, and pounding tools (BarYosef 2001). At this time there is a greater occurrence of a burial practice in which the skull is separated from the body (Rollefson 2003). In particular, the Sultanian Industry has evidence of this practice with single burials with no grave goods with the skulls recovered from domestic areas (Bar-Yosef 2001). Symbolic representations come in the form of limestone or clay anthropomorphic figurines interpreted possibly as females standing or kneeling (Bar-Yosef 2001). Evidence for long distance trade during the PPNA comes from obsidian imported from central Anatolia and marine shells from the Red Sea (Bar-Yosef 2001). During this time there is evidence of wild cereal and pulse cultivation (Colledge 1998; Willcox et al. 2008; Willcox 2011) accompanied by the hunting of gazelle, equids, and cattle in the middle Euphrates and gazelle, fox, fallow deer, wild boar, and wild cattle in the Jordon Valley (Bar-Yosef 2001). PRE-POTTERY NEOLITHIC A OF CYPRUS The Cypro-PPNA (Manning et al. 2010; McCartney 2005; McCartney et al. 2007) represents the first migration of semi-permanent complex foragers from the mainland Levant to Cyprus. The migration involved mainland Late PPNA groups and possibly with inhabitants of sites dated to the earliest PPNB levels on the Euphrates (Manning et al. 2010). The Cypriot PPNA coincides with more warm and humid conditions characteristic of the beginning of the Holocene (Rosen 2007) and thus, climatic conditions possibly made Cyprus more attractive (Manning et al. 2010).

PRE-POTTERY NEOLITHIC B OF THE MAINLAND LEVANT AND CYPRUS

Agia Varvara-Asprokremmos is one of a handful of sites that have been recently surveyed and excavated by the Elaborating Early Neolithic Cyprus (hereafter EENC) project (directed by Carole McCartney and Sturt Manning). The aim of this project is to explore the whereabouts of presumed early hunter-gatherer sites that have been unrepresented due to past survey and excavation biases that have focused on single

PRE-POTTERY NEOLITHIC B OF THE MAINLAND LEVANT Multiple significant cultural changes occurred during the PPNB on the mainland, including changes in architecture, technology, burial practices, and trade, which were accompanied by population increase, expansion and, of course, population movement. The 14

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT PPNB of the mainland Levant is divided into four phases, Early, Middle, Late, and Final. The PPNB emerges in the Northern Levant and it subsequently spread to other regions (Bar-Yosef 2001). Sites dated to the PPNB and that are located in the northern Levant include Jerf al-Ahmar (Stordeur 2003), Mureybet (Cauvin 2000), Çayönü (Özdoğan 1999), and Nevali Çori (Hauptmann 1999) and sites that are located in the southern Levant include Jericho (Kenyon and Holland 1982) and ‘Ain Ghazal (Rollefson 1993).

Architectural features of the Cypro-EPPNB include wells, pits, and post-hole alignments. For example, at Parekklisha-Shillourokambos a large trapezoidal enclosure was recovered and was presumably used for keeping livestock (Guilaine and Briois 2001; Peltenburg et al. 2003). The wells at Parekklisha-Shillourokambos (Early Phase A) and Kissonerga-Mylouthkia (Period 1A) are the earliest known water-wells and have become a diagnostic feature of the early colonizing sites on the island (Peltenburg et al. 2001, 2003). Another characteristic of the early colonizing sites is the relatively high occurrence of imported Anatolian obsidian, which further supports external links with the mainland at this time (Peltenburg et al. 2003; Guilaine and Briois 2001). The chipped stone industry demonstrates the use of high quality translucent chert in the manufacture of a blade based industry of projectile points and sickles, the latter being used in the harvesting of cereals (Guilaine and Briois 2001; Peltenburg et al. 2001). The ground stone industry includes predominately stone vessel fragments and crude limestone hammer stones as well as axes, pounders, grooved stones, and mace heads (Peltenburg et al. 2003). The occurrence of cutting tools are rare and artifacts for food processing, including querns and rubbers, are absent from the earliest level at Kissonerga-Mylouthkia (Peltenburg et al. 2003).

There is a change in domestic architecture from the proceeding mainland PPNA, including a change from curvilinear domestic structures to rectangular buildings with multiple lime-stone plastered floors (Bar-Yosef 2001; Rollefson 2003). The chipped-stone industry comprises arrowheads and sickle blades that were constructed on the bi-directional naviform core technique (Bar-Yosef 2001; Rollefson 2003). Human burial practices during the PPNB are suggestive of social hierarchy with a dichotomy in the treatment of the dead. For example, intact bodies have been found apparently discarded in secondary deposits. Further, in the Levant the decapitated skulls were treated with plaster and pigment, in particular at ‘Ain Ghazal (Rollefson 1993). In the MPPNB there appears to be a greater representation of wild cattle figurines and human representations representing female fertility (Rollefson 2003).

The Cypro-Middle PPNB is a period of amalgamation and contact with the mainland and is followed by a development phase in the Cypro-LPPNB, which is marked by a supposed decrease in external contact (Peltenburg 2004a, 72). Cypro-MPPNB occupation is limited to Parekklisha-Shillourokambos (Middle) and the sites with Cypro-LPPNB occupation include KissonergaMylouthkia (Period 1B), Kalavasos-Tenta (2-4), Parekklisha-Shillourokambos (Middle and Late), and Krittou Marottou-‘Ais Yiorkis (Middle/Late) (Peltenburg et al. 2001; Peltenburg et al. 2003). Architecture includes wells (Kissonerga-Mylouthkia), pits, circular mud-brick domestic structures, hearths (Peltenburg et al. 2003) and a mud-brick and stone wall enclosure at Kalavasos-Tenta (Todd 2001; 2003). An insular adaptation involving unidirectional core technology based on a local opaque material continues into the subsequent Khirokitian, replacing the bi-directional core reduction using high quality translucent chert (Peltenburg et al. 2003; Peltenburg et al. 2001). The ground stone assemblage consists of mace heads, pierced limestone discs, stone bowl fragments, hammer stones, hole-mouth vessels, pounders and anvils (Peltenburg et al. 2003). Similar to the preceding Cypro-EPPNB, cutting tools are rare and querns and rubbers are absent (Peltenburg et al. 2003). There are parallels in burial practices with preceding mainland traditions, particularly the practice of multiple secondary burials and post-mortem removal of the skull from the body, which is evident at KissonergaMylouthkia (well 133) (Peltenburg et al. 2003). Mainland contact during this phase comes from evidence of imported south-central Anatolian (Çiftlik) obsidian (Todd 2003; Simmons 2003) and external influence is also apparent in the form of anthropogenic wall

PRE-POTTERY NEOLITHIC B OF CYPRUS It is during the PPNB that farming communities spread from the mainland to other regions including Cyprus. In regards to the spread of the PPNB Finlayson (2004, 1920) states that at this time there are notable differences in site types, architecture, decoration, environmental location, and economies on the mainland but overall a single PPNB ‘multifaceted phenomenon’ is evident. Further, the evidence from early sites on Cyprus places the island in the mainstream tradition with its unique variations adapted to the local environment. The CyproEarly PPNB provides the first evidence of a Levantine migration of an agricultural group in the mid-late ninth millennium BC (Peltenburg et al. 2001; 2003; Peltenburg 2004a). Peltenburg et al. (2000) argue for strong analogies with the northern Levant (Peltenburg et al. 2000; 2001; 2003), including similarities with sites in the mid-upper Euphrates such as Jerf el Ahmar, Abu Hureyra 1 and 2, Göbekli Tepe, Halula, Çayönü, and Mureybet (Peltenburg et al. 2003, 95). In opposition Simmons (Simmons 2004, 11) observes parallels with the southern Levant including Jericho, ‘Ain Ghazal, Wadi Shu’eib, and Gwair I, based on the similarity of settlement types. Also, Colledge el al. (2004; see also Colledge and Conolly 2007) suggest origins in the southern Levant for the crop and weed assemblages represented in the charred macro remains found at the sites. This issue will be discussed further in Chapter 7, where multiple origins are suggested (Lucas et al. 2012).

15

CROPS, CULTURE, AND CONTACT paintings, similar to those at Middle PPNB Halula and Çatal Höyük, found at Kalavasos-Tenta (Todd 2003; Peltenburg et al. 2003).

75). This architecture illustrates a dichotomy with the rectilinear buildings of contemporary mainland sites, which persist from this date forward. However, the circular architecture of Cyprus demonstrates parallels with the architecture of earlier mainland occupations, specifically at Late Mureybetian/PPNA Mureybet III, Jerf el Ahmar and Munhatta 3 (Peltenburg 2004b, 79).

It is during the 7th millennium BC that the mainland PPNB culture is said to have collapsed (Akkermans and Schwartz 2003; Bar-Yosef 2001; Kuijt and GoringMorris 2002; Rosen 2007, 37; also see Maher et al. 2011 for alternative argument). Evidence for this cultural collapse includes site abandonment and the replacement of large villages with smaller communities (Bar-Yosef 2001). However, there is evidence of population increase at some sites including Abu Hureyra, Basta, ‘Ain Ghazal, and Wadi Shu’eib (Rollefson 2003). The PPNC culture is characterized by a decrease in the use of bidirectional naviform technology, grinding stones and sickle blades, in human and animal symbolic representations, and in the ritual of skull detachment (Rollefson 2003).

The chipped stone of this phase has been described as rough and impoverished and lacks variation, pressure retouch, and specific tools such as arrowheads (Le Brun 2001). The ground stone industry includes stone vessels, mace heads, incised stones, batons, figurines, and axes, which it is suggested are indicative of larger scale clearance of woodland in advance of cereal crop agriculture (Peltenburg et al. 2003; Steel 2004). Changes in funerary practices are evident with single primary pit burials replacing the multiple secondary burials of the previous Cypro-PPNB (Peltenburg et al. 2003; Le Brun 2001). A decrease in the incidence of imported Anatolian obsidian suggests a decrease in contact with the mainland during this period (Le Brun 2001; Peltenburg et al. 2000). However, on the basis that it would have been necessary to continually re-introduce fauna to the island in order to sustain island populations, Horwitz et al. (2004) argue for continuous contact between the island and the mainland at this time. Additionally, McCartney and Gratuze (in Peltenburg et al. 2003) argue for sustained membership in the PPNB interaction sphere based on similarities in the chipped stone assemblage. Moreover, Peltenburg (2004b, 83) argues that the continued use of circular domestic architecture is an adaptive response to low population densities, the island’s limited resources, and a lack of inter-group competition. Thus, the level of external contact during this time did not necessarily decrease and the differences evident between the island and the Levantine/Anatolian mainland could be a result of an insular adaptation as opposed to isolation from mainland populations.

KHIROKITIAN Le Brun (2001, 33-34) states that from the second half of the 7th millennium BC, Cyprus experienced internal cultural evolution and developed a unique Cypriot identity distinct from contemporary mainland cultural traditions and which reached full expression in the late Aceramic Neolithic, or the Khirokitian Culture. It is from this point forward that Cyprus veered from the mainland cultural trajectory not to return to it until the Bronze Age. The Khirokitian culture of Cyprus is contemporary with the mainland Pottery Neolithic and differs from mainland traditions in multiple ways, of great significance is pottery manufacture, architecture and social complexity. Whilst the mainland cultures were experimenting with unfired pottery in the Late PPNB in the northern Levant (Syria and Jordan) and hand-made fired pottery in the Pottery Neolithic (i.e. pre-Halaf and Halaf, c. 6570-5400 cal. BC) (Bar-Yosef 2001; Rollefson 2003), the distinct cultural identity of the last phase of the Aceramic Neolithic of Cyprus, the Cypriot Khirokitian (ca. 70005200 cal. BC), was only just emerging. Although there is evidence for experimentation with early pottery technology, including roughly made, partly fired, (perhaps cooked), clay vessels and unbaked figurines in the Cypriot Aceramic Neolithic, there is no evidence for firing technology that is required for pottery production (Clarke 2007, 11).

CERAMIC NEOLITHIC OF CYPRUS AND THE CHALCOLITHIC OF CYPRUS AND THE MAINLAND LEVANT CERAMIC NEOLITHIC OF CYPRUS The Ceramic Neolithic of Cyprus (i.e. Sotira Culture) dates between 4700-3900 BC (Clarke 2001; Clarke et al. 2007) and roughly corresponds to the mainland Chalcolithic, which dates to ca. 4500-3600 (Burton and Levy 2001; Joffe and Dessel 1995). The sites dated to the proceeding Khirokitian culture are abandoned at ca. 5500 cal. BC and there is a chronological gap between the Khirokitian culture and subsequent Sotira culture. There is debate over whether the gap in the archaeological record is one of true site-abandonment or rather is a result of lowered site visibility due to increased population mobility as a result of greater emphasis on hunting (Peltenburg in Hadjisavvas 2010). The sites with Ceramic Neolithic occupation include Ayios EpiktitosVrysi, Philia, Dhali-Agridhi, Sotira, Kalavasos-Tenta and Klepini-Troulli, Kantou, Khirokitia-Vounoi, and

The Khirokitian cultural entity is considered to be the pinnacle of the Aceramic Neolithic culture in Cyprus (Peltenburg 2004b, 72). Evidence comes primarily from the largest site of the Aceramic Neolithic, KhirokitiaVounoi. Other sites that have Khirokitian occupation include Dhali-Agridhi, Cape Andreas-Kastros, Kalavasos-Tenta, Kholetria-Ortos and KissonergaMosphilia. The major archaeological feature of this phase is the thick-walled circular domestic structure, specifically the circular pillar (CPB, buildings with internal large rectilinear pillars) and circular radial buildings (CRB, buildings with no central installations and peripheral cells or partitions) (Peltenburg 2004b, 7316

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT Kokkinoyia (Peltenburg 1979b; Clarke et al. 2007; Clarke 2001). The architecture of this phase includes free-standing, mono-cellular, rectilinear stone and mud brick semi-subterranean structures, which include partition walls, hearths, and pisé and stone benches (Clarke et al. 2007; Peltenburg 1978). Changes in architecture and the organization of domestic space from the preceding phase are evident, with regularization in the arrangement of domestic fireplaces, benches and work craft activity areas (Peltenburg 1993). However, continuity from the preceding Khirokitian is demonstrated with parallels in chipped stone and ground stone assemblages (Clarke et al. 2007) despite the considerable time lapse between the periods. The ground stone assemblage of this period includes stone bowls, grinders, hammer stones, pestles, and mortars (Clarke et al. 2007; Clarke 2001, 67; Steel 2004, 63-81). The material culture of the Ceramic Neolithic is similar at all sites (Peltenburg 1993; Clarke et al. 2007) and includes limestone vessels, picrolite and bone ornaments, axes, adzes, chisels, and pierced discs. McCartney (in Clarke et al. 2007, 90) states that “given the link suggested by the engraved stone pebbles of Khirokitia-Vounoi and Kholetria-Ortos and those of Yarmoukian sites (from Byblos to Sha’ar Hagolan) it seems highly likely that Cyprus participated in this sphere of interaction, continuing this relationship with the central and southern Levant during the late Neolithic period” (citing BarYosef 1992; Eirikh-Rose 2004; Garfinkel 2004).

Middle Phases of the Ceramic Neolithic (Peltenburg, 1982, 39). CHALCOLITHIC OF THE MAINLAND LEVANT Evidence for human occupation on the island at this time is considered ephemeral. In opposition, the mainland was undergoing a ‘technological’ or ‘specialisation revolution’ (Maisels 1999). This revolution includes the establishment of temples and burial grounds, the emergence of craft specialisation, specifically an increase in the production of copper, ceramic and stone tools, and changes in the mode of interaction and trade with evidence for pack animals (i.e. donkey) (Maisels 1999; Rowan and Golden 2009). In contrast to the Ceramic Neolithic of Cyprus, contemporary Chalcolithic communities on the mainland were undergoing population increase, settlement expansion, and technological and economic developments, including aspects of the ‘secondary products revolution’ (discussed below) (Maisels 1999; Sherratt 1981). In addition to temples and cemeteries, there is evidence of large rectangular mud brick domestic structures with stone foundations, lime-plastered floors, and adjoining courtyards (Maisels 1999). The chipped stone assemblage includes scrapers, sickle-blades, retouched and backed blades, retouched bladelets, notches, denticulates, awls, borers, bi-faces (i.e. axes, adzes, and chisels), burins, arrowheads, and hammer stones (Levy 1986, 90). There were advances in ceramic technology, including evidence for wheel-made (i.e. slow wheel or tournette) pottery, with the inclusion of the following vessel types: v-shaped bowls, large pithoi, hole mouth vessels, small globular jars, jars, bowls, basins, footed vessels, vessels with multiple handles, and churns (Levy 1986). Joffe and Dessel (1995, 514) have described the ‘Late Developed Chalcolithic’ (ca. 3900/3800-3700) as a period when Chalcolithic society “reaches its height of expansion, in terms of geographic extent, size and density of sites, the intensity of agropastoral production, and the complexity of procurement, production, and exchange networks.” The “Late Developed Chalcolithic” corresponds with the beginning of the Cypriot Chalcolithic, discussed below.

The first evidence of pottery in Cyprus comes from three sites: Dhali-Agridhi in the central Mesaoria, Philia in the Morphou Bay and Klepini-Troulli on the north coast, and is characterized by two types: a common Coarse Ware and a Dark Burnished Ware (Clarke et al. 2007). The origin of the Cypriot ceramic technology is assumed to have come from the mainland since there is no evidence on the island of pyro-technology prior to the Ceramic Neolithic (Clarke et al. 2007). Clarke et al. (2007, 92) argue a mainland influence based on the following similarities: coil and slab method construction, firing techniques, the decoration of red paint, use of mats and basketry in vessel construction, the limited range of shapes, and the manufacture of coarse types. The early pottery production of Cyprus demonstrates regional homogeneity in regards to manufacturing techniques and morphology (Clarke 2007; Clarke et al. 2007). It was manufactured at the household level, fired in open hearths and includes the use of local clay resources in the production of simple-coil, hand-made vessels (Clarke 2007; 2001). The ceramic types include hemispherical bowls, ovoid jugs, bottles and hole-mouth jars. Variation in decoration has been linked with regional group identity, with the northern sites expressing identity through Broad Line Red on White styles and the Southern sites with the Combed Ware styles (Peltenburg 1982c, 40; Clarke 2001). Moreover, the Broad Line Red on White from Ayios Epiktitos-Vrysi is a contemporary regional variant of the Combed Ware of Sotira, suggesting regionalism in the ceramic repertoire in the

CHALCOLITHIC OF CYPRUS AND THE EARLY AND MIDDLE BRONZE AGE OF THE MAINLAND LEVANT THE EARLY BRONZE AGE OF THE MAINLAND LEVANT The Cypriot Middle Chalcolithic (ca. 3500-2800 BC) is contemporary with the Early Bronze Age (ca. 3500-2000 BC) on the mainland. Although there is continuity with the earlier Chalcolithic, there were a lot of cultural and technological changes of the mainland Bronze Age culture, including the emergence of the urban state (ca. 3100 BC) accompanied by changes in domestic and funerary architecture, the use of public space, craftspecialisation (e.g. metalworking), widening international trade networks (including Egypt), and agricultural modes and technology, including the first 17

CROPS, CULTURE, AND CONTACT unequivocal evidence for grape and olive cultivation, the cattle-drawn plough, and pack animals (i.e. donkeys) (Genz 2000; Richard 1987). Architecture on the mainland at this time includes urban states with paved streets, multi-room rectangular houses with courtyards and large-scale storage, centrally located temples, and fortification walls (Richard 1987). Pottery types include inverted-rim bowls, ‘teapots’, four-spouted lamps, and ledge-handled jars (Richard 1987). Metalworking included the use of copper, silver, gold, and tin for the manufacture of tools, weapons, and jewellery (Genz 2000).

natural resources of the mineral, which outcrops from the Kouris and Karyotis rivers. These figurines have been found in direct association with the characteristic Chalcolithic deep, bell-shaped shaft burials (Peltenburg et al. 2006; Xenophontos 1991). The first attempts of copper metallurgy are seen with a copper hook recovered from Kissonerga-Mylouthkia, a chisel tip and hook from Erimi, a chisel and possible blade from Lemba, and a snake ornament from the Souskiou-Laona settlement (Crewe et al. 2005; Gale 1991; Peltenburg et al. 1998; 2006). Evidence for intense external contact in the Late Chalcolithic of Cyprus comes from the Pithos House from Kissonerga-Mosphilia (Period 4B). The basis for this is the evidence of new tool types, including crucibles and spindle whorls for textiles, new pottery styles, such as Red Polished jugs, large serving bowls, bulk storage vessels (i.e. pithoi), and olive oil production, all of which suggest contact with Anatolia during this time (Steel 2004; Peltenburg et al. 1998). There is an increase from the preceding periods in the quantity of axes and adzes, which have been used to infer greater labour investment in timber-cutting for land clearance and woodworking (Peltenburg et al. 1998). The cemeteries of the Middle Chalcolithic appear to be abandoned by the Late Chalcolithic, as are the associated picrolite funerary figurines. The Red on White pottery of the preceding phases (Lemba 2/Kissonerga 3B) declines and is replaced by the Red and Black Stroke-Burnished ware of the Late Chalcolithic (Peltenburg et al. 1998; Peltenburg et al. 2006). Evidence of experimentation with clays, slips, fire temperature and fire control suggests a shift from production of pottery at the household level to more specialized production (Bolger and Shiels 2003, 168). Evidence of Cypriot contact in the Late Chalcolithic off the island comes from two types of Cypriot Chalcolithic pottery recovered from Anatolian Tarsus (Frankel and Webb 2006, 104). Archaeological evidence in Cyprus of Anatolian contact comes from imported obsidian, which had been absent since Aceramic Neolithic levels.

CHALCOLITHIC OF CYPRUS Peltenburg states (2010, 51) that the Chalcolithic of Cyprus (i.e. Erimi culture) is characterized by significant population growth and innovations in art (e.g. symbolic representations), craft production, metallurgy and the first signs of social inequality and intensification of ritual and economy. The Chalcolithic of Cyprus dates from ca. 3900 to 2400 BC and is sub-divided into an Early, Middle and Late occupation. Many of the sites on Cyprus dated to the Sotira culture were abandoned, possibly due to climatic instability, and the Erimi culture was established due to population reorganization in the earliest phases (Peltenburg in Hadjisavvas 2010). Peltenburg et al. (1998) have described the Early and Middle Chalcolithic (Kissonerga-Mosphilia Period 4A) as pre-Anatolian contact and the Late Chalcolithic as a period of increasing external contact and subsequent development (Kissonerga-Mosphilia Period 4B-5). At this time continual external contact begins and the island goes from relative isolation and independence to involvement in the broader Mediterranean interaction sphere (Peltenburg et al. 1985). Sites with Chalcolithic occupation include Erimi-Pamboula, the cemeteries and settlement of the Souskiou complex, KissonergaMosphilia, Kissonerga-Mylouthkia, Lemba-Lakkous, and Kalavasos-Ayious (Peltenburg et al. 2006; 1985; Todd 1991). The Chalcolithic of Cyprus is marked by multiple cultural changes including new forms of domestic and funerary architecture, ideology in the form of new symbolic art, and the beginnings of copper metallurgy (Gale 1991; Peltenburg et al. 1985; Peltenburg 1991). Domestic architecture of this period consists of freestanding, timber-framed semi-subterranean circular pisé structures with flat roofs, stone foundations, plastered walls, centralized hearths, and partition ridges for household organization (Frankel 2005; Peltenburg et al. 2006; Steel 2004; Thomas 2005). The ground stone assemblage consists of axes, adzes, anvils, chisels, hammer stones, pestles, pounders, querns, stone vessels, and rubbers (Todd 1991; Peltenburg et al. 1985; Peltenburg et al. 2003; Peltenburg et al. 2006; Peltenburg et al. 1998). The pottery consists of flasks, bowls, jars, goblets, bottles, and anthropomorphic and zoomorphic vessels (Peltenburg et al. 2006). Cruciform picrolite figurines represent early exploitation of the island’s

EARLY AND MIDDLE CYPRIOT BRONZE AGE The Bronze Age of Cyprus has been sub-divided into the following phases: Philia, Early Cypriot I-III, and Middle Cypriot I-III, with the Early Cypriot cultural complex replacing Philia sometime around 2,300 BC (Frankel and Webb 2006a, 307). This terminology and chronological schema is used here as opposed to that proposed by Knapp (2008), which places the Late Chalcolithic as PreBronze Age, the Philia and Early Cypriot I-II as PreBronze Age 1, and the Early Cypriot III-Middle Cypriot I-II as Pre-Bronze 2. At this time there is a dramatic rise in population with an estimation of about 100 sites in the Chalcolithic versus over 300 in the Early and Middle Cypriot phases (Swiny 2008). Sites occupied during the Philia phase include Marki-Alonia and KissonergaMosphilia and settlements representing the later, Early/Middle Cypriot phase include EpiskopiPhaneromeni, Marki-Alonia, and Sotira-Kaminoudhia. 18

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT The phase of increased external contact and possible initial settlement of Anatolian migrants in the Late Chalcolithic was followed by one of consolidation and stabilization. This is described archaeologically as the Philia facies of the Cypriot Early Bronze Age (Frankel and Webb 2006; Frankel 2005; Webb and Frankel 1999). Although most of the information on Philia comes from cemeteries, there are two sites with settlement evidence, Kissonerga-Mosphilia (Period 5) and Marki-Alonia (Phase A-B) (Peltenburg et al. 1998; Frankel and Webb 2006). Absolute dates are limited for this phase but the Philia occupation at Marki-Alonia (Phase A-B) suggests a range from about 2400 to 2200 BC (Frankel and Webb 2006a) with evidence for a possible migration from southwest Anatolia (Frankel 2005, 19) at about 2400 BC (Webb 2001).

settlements (Webb 2001; Swiny 2008; Frankel 2005). The ground stone assemblage includes mortars, querns, rubbing, and gaming stones (made from igneous rocks) (Frankel and Webb 2006; Swiny 2008), with mortars, basins, rubbers, and querns occurring much more than in previous phases (Frankel and Webb 2006a). There is, however, a marked absence in the occurrence of stone vessels (Frankel and Webb 2006a). New forms of food preparation include cooking in pots and baking in pans and ovens (Webb 2001). Changes in weaving techniques and loom construction involved the use of low-whorl spindles and clay weights for warp-weighted looms (Webb 2001; Frankel 2005). The abstract ‘plank-shaped’ female figures appear at this time demonstrating new forms of symbolic art (Frankel 2005). Increased numbers of copper artifacts of more complex design, including spearheads, daggers, axes, razors, tweezers, awls, rings, earrings, and bracelets provide evidence both of technological advances in ore extraction and production (Swiny 2008; Webb 2001).

The Philia Phase is transitional between the Late Chalcolithic and the Early Bronze Age of Cyprus (Knapp et al. 1990; Webb and Frankel 1999). It is a distinct cultural phase and in all aspects represents change from the preceding period, for example, in the style of domestic architecture, the mode of agricultural production, food preparation, drink, dress, and burial practices (Peltenburg 1996). More specifically, during the Early Bronze Age there were changes with the introduction of ore extraction and copper production and plough-based agriculture. Also, the circular-based domestic architecture of the preceding Aceramic Neolithic, Ceramic Neolithic, and Chalcolithic is replaced by multi-cellular rectilinear structures. There were also changes in burial practices with evidence of individual chamber tombs and extramural cemeteries (Knapp 2008; Frankel and Webb 2006). Changes in domestic technologies include the occurrence of the vertical warp-weighted loom, the low-whorl for spinning, and new ceramic forms, including vessels suitable for containing boiled liquids for cooking and serving vessels, including small bowls, cutaway-spouted jugs, juglets, small jars, amphorae, and flasks. The characteristic pottery type is the Red Polished Philia Ware (Bolger 1991; Frankel and Webb 2006).

Improvements in metallurgical techniques, in particular, played an important role in increasing external contact and possible migration in the Early Bronze Age of Cyprus. It is argued that an elite northern group living near the copper outcrops (Peltenburg 1996, 27) advertised to Anatolia the island’s potential economic resource, which resulted in trade interest with Anatolia (Manning 1993, 35). Debates continue regarding the level of external involvement and contact: some have argued for an indigenous emergent copper-producing group that participated in the overseas copper trade (Knapp 1993; Knapp et al. 1990; Manning 1993), whilst others have argued that Anatolians migrated to the island to exploit the copper resources (Frankel et al. 1996; Frankel and Webb 2006). Both arguments are well supported and possible. It is likely that increasing external contact with limited migration began in the Late Chalcolithic and increased over time, which subsequently lead to the development of the Late Cypriot urban society. Although excavations at Marki-Alonia, Sotira-Kaminoudhia, and Episkopi-Phaneromeni provide evidence for settlements in the Early and Middle Cypriot periods, more data are needed to better understand the relationship between external interaction, migration, and the development of the Late Cypriot Bronze Age.

Intensification in agricultural practices has been inferred from a significant increase in the occurrence of ground and chipped stone tools used in the processing of cereal crops (Frankel and Webb 2006; Knapp 2008; Webb 2001). The re-introduction of cattle to the island could be the most significant development of the prehistoric Bronze Age (Swiny 2008, 43) because it provided a means for agricultural extensification through plough based agriculture in addition to new source of milk production (i.e. cow’s milk) (Knapp 1990; Swiny 2008; Steel 2004).

SUMMARY SUBSISTENCE ON PREHISTORIC CYPRUS A more detailed discussion of the plant-based subsistence of the island will be presented in Chapter 4 but the faunal evidence will be introduced briefly here. The mammalian fauna of Cyprus is limited and all species either travelled to the island by sea prior to human migration or were introduced by humans, e.g., from the mainland Levant. The species that were on the island prior to human occupation include the Cypriot pygmy hippopotamus (Phanourios minutus) and the pygmy elephant (Elephas cypriotes), both of which are small due to an island adaptation (i.e. insular dwarfism) (Simmons 1999; Horwitz et al. 2004; Vigne 2011). Wild

The Early and Middle Cypriot periods will be discussed together here to provide a general overview of the cultural complexes. The architecture of these phases includes multi-roomed rectilinear mold-made, mud-brick domestic structures with rectangular hearths, low benches, and cemeteries located away from domestic 19

CROPS, CULTURE, AND CONTACT pigs were introduced from the mainland during the Akrotiri phase and again during the PPNA at Asprokremmos and suggest some form of pig management or incipient domestication (Vigne 2011). Vigne (2011) suggests that the pigs were introduced to the island sometime during the mainland Natufian (Late Glacial), decreased in size, and then were hunted by visitors from the mainland Levant (evidence from Ayia Varvara-Aetokremnos). By the latest Aceramic Neolithic phase at Parekklisha-Shillourokambos and KissonergaMylouthkia all species that represent the characteristic Cypriot archaeological assemblage are present, including sheep, goat, pig, cattle, fallow deer, dog, fox, genet, the unintentionally introduced house mouse, and cat (Vigne et al. 2000; Horwitz et al. 2004; Peltenburg et al. 2003) (for list of faunal introductions and assemblages for each cultural phase refer to Tables 2.2 and 2.3). There is debate regarding the domestication status of the introduced species at the time of import. The argument depends on how much significance is placed on either morphological (Horwitz et al. 2004) or nonmorphological domestication criteria (Zeder 2008). Domestication criteria based on morphology include a reduction in body size, bone density, and horn form, whereas those based on non-morphological criteria include a shift in age and sex ratios of culled animals, and an increase in targeted species (Horwitz et al. 2004; Vigne 2011).

production. Thus, the evidence of goats at ParekklishaShillourokambos provides the first evidence of a domestication process on a Mediterranean island with a new linage of domestic goat appearing at ca. 9400-9000 BP (Vigne 2011). Subsequent to goat introduction, small, horn-modified domestic sheep were introduced. Cattle were bred and are present in the earliest levels of Parekklisha-Shillourokambos, Akantou in the north, and Middle/Late Cypro-PPNB Krittou Marottou-‘Ais Yiorkis but disappears by the Khirokitian, not to return until the Early Bronze Age when they are re-introduced as a domestic species (Vigne et al. 2000; Simmons 1998; Croft 1991; Sevketoglu 2000). It has been suggested by Horwitz et al. (2004, 38) that the disappearance of cattle could be due to a lack of introduced fresh stock for the maintenance of the island’s population. The controlled exploitation and hunting of Mesopotamian fallow deer, which is a unique island adaptation, begins in the Aceramic Neolithic, increases in the Ceramic Neolithic and Chalcolithic, and decreases in the Middle and Late Chalcolithic (Croft 1991). Further, it is the primary meat source for most of the sites until the Late Chalcolithic (Croft 1991; Legge 1982; Vigne and Buitenhius 1999; Peltenburg et al. 2000; Vigne et al. 2000; Horwitz et al.. 2004). Of note, is the evidence of a cat in a human burial at Parekklisha-Shillourokambos dating to ca. 7300-7200 BC, which is suggestive of early taming and potential domestication of the animal prior to evidence of domestication in Egypt in the 2nd millennium BC (Vigne et al. 2004). Evidence of domesticated cat (and dog) has also been demonstrated earlier in the Cypro-PPNA from the recently excavated Ayios Tychonas-Klimonas (Vigne et al. 2011b). In addition to the re-introduction of domestic cattle in the Early Bronze Age, there was an introduction of equids (i.e. donkeys) (Croft 1996, 1996; Horwitz et al. 2004; Vigne 1999). It is clear that the faunal record of early Cyprus is complex, with different species at different stages of management and domestication introduced to the island at different times.

Morphological criteria suggest multiple introductions of wild populations due to the size of the Cypriot bones, which are large and robust (Horwitz et al. 2004). The exception to this is pig, which is the first mammal introduced to Cyprus and appears in assemblages from Akrotiri and Ayia Varvara-Aetokremnos (Horwitz et al. 2004; Simmons 1999; McCartney et al. 2007; Vigne et al. 2000). Horwitz et al. (2004) argue the small size could be the result of selection for an easier transport. In opposition, scholars that place weight on nonmorphological criteria argue for introductions of wild/managed animals that can be considered predomestic animals ( Clarke et al. 2007; Vigne et al. 2009; Vigne 2011). Based on evidence from Akrotiri, Vigne et al. (2009) argue that wild pig populations were managed before they were introduced to Cyprus and thus could be considered pre-domesticated. Recent evidence from Parekklisha-Shillourokambos supports a model that demonstrates a long time span of increasing intensive control of wild boar populations, which indicated that wild boar management could have occurred more than 11,400 years ago (Vigne 2011). Additionally, Vigne (2011) argues for an introduction of pre-domesticated populations of goat and sheep, with evidence from Parekklisha-Shillourokambos suggestive of multiple introductions of different domestic sheep, pig and cattle lineages in the course of the tenth millennium (Vigne 2011). The evidence from Parekklisha-Shillourokambos suggests hunting of goats in the earliest phases, more intensive exploitation in the middle phases, and morphological changes evident in the late phases, with modifications of culling profiles suggestive of milk 20

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT

Figure 2.2 Map of Cyprus showing modern average rainfall distribution (recreated by Chris Stevens after Zohary 1973, 29)

Figure 2.3 Map of Cyprus showing the sites analysed in this thesis and the sites mentioned in text

21

CROPS, CULTURE, AND CONTACT

Akrotiri

Cypro-PPNA

Cypro-PPNB

Khirokitian

Notable Innovations or Introductions

Exploratory Phase

Exploratory Phase with evidence for hunterforager occupations

Migration of crop-based farmers from the mainland

Decrease in Anatolian obsidian

Agriculture

No evidence

No evidence

Migration of crop-based farmers

Hoe-based primary products

Pig, dwarf elephants and hippos, genets, mice, birds, marine invertebrates

Pig Fresh water crab fish

Sheep, goat, fallow deer, pig, cattle house mouse, dog, cat, fox, fish, fresh water crab

Sheep, goat, fallow deer, pig, cattle, house mouse, dog, cat, fox, fish remains

Faunal

Chipped stone

Ground stone

Thumbnail scrapers, burins, retouched blades, bladelets, flakes

No evidence

Core reduction

High quality translucent chert, blade-based, projectile points and sickles, unidirectional core technology by Khirokitian

Rough industry, pressure retouch, arrowheads, platform core technology

Grinders, rubbing stones, pestles, quern fragments, pounders, hammer stones

Stone bowl vessels, hammer stones, axes, flaked tools, pounders, grooved stones and mace heads, limestone discs, anvils

Stone vessels, mace heads, incised stones, batons, figurines, and axes

Chipped stone similar to N. Levant

Anatolian obsidian declines by LPPNB, secondary burials and skull removal, Plant and animal assemblages. Wall paintings

Architectural parallels with earlier mainland phases, decrease in Anatolian obsidian

Evidence of external contact

Exploratory

Additional artefacts

Stone and shell beads

Shaft straightener

Architecture

Hearths

pits

Wells, pits, hearths Mud-brick curvilinear structures, post-hole alignments, enclosures

Heavy-walled circular domestic structures

Single primary pit burials beneath domestic structure floors

shells, and bone fishhooks

Burial

No evidence

No evidence

Secondary burials in association with mace heads, caprine carcasses and post-mortem skull removal

Textile production

No evidence

No evidence

No evidence

No evidence

Ceramic

No evidence

No evidence

No evidence

No evidence

Settlement distribution

Little evidence

Little evidence

Concentration on richer, better watered coastal regions

Metallurgy

No evidence

No evidence

No evidence

Concentration on richer, better watered coastal regions

No evidence

Table 2.2 Summary of differences between early Cypriot cultural phases (data taken from Webb 2001; Horwitz et al. 2004; Frankel 2005, 21; Simmons 2004; McCartney et al. 2006; Guilaine and Briois 2001; Peltenburg et al. 2003; Todd 2003;Le Brun 2001; Clarke et al. 2007; Peltenburg 1978; Clarke et al. 2007; Clarke 2001; Steel 2004;Todd 1991; Peltenburg et al. 1985; Peltenburg et al. 2003; Peltenburg et al. 2006; Peltenburg et al. 1998; Manning 1993; Peltenburg 1996; Frankel et al 1996; Frankel and Webb 2006; Swiny 2008)

22

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL BACKGROUND OF CYPRUS AND THE MAINLAND LEVANT

Ceramic Neolithic

Chalcolithic

Philia/Early Cypriot Settlement Shift, recti-linear architecture, introduction of the plough, re-introduction of cattle, introduction of the donkey, significant increase in crop processing tools, and new forms of food production consumption technologies Plough-based; backed sickles Secondary Products

Notable Innovations or Introductions

First evidence of pottery

First evidence of olive oil production, bulk storage, return of imported Anatolian obsidian, first evidence of spindle whorls, no direct-fire boiling pots, and new pottery types

Agriculture

Hoe-based primary products

Hoe-based primary products LChal: Olive oil production, bulk storage

Faunal

Sheep, goat, fallow deer, pig, house mouse, dog, cat, fox

Sheep, goat, pig, deer

Re-introduction of Cattle, sheep, goat, pig, deer, introduction of donkeys and mustelids

Chipped stone

Continuity from Khirokitian, limestone vessels, axes, adzes, chisels, pierced discs

Long blades, burins, denticulates, scrapers, glossed blades, sickles

Significant increase in chipped stone

Ground stone

Continuity from Khirokitian, stone bowls, grinders, hammer stones, pestles, and mortars

axes, adzes, anvils, chisels, hammer stones, pestles, pounders, querns, vessels, and rubbers LChal: increase in axes and adzes

Significant increase crop processing: mortars, querns, large clay basins, rubbing and gaming stones NO stone vessels

Evidence of external contact

Evidence of pottery/firing techniques/red paint decoration,, coil and slab method construction,

LChal increasing contact, imported obsidian

New forms of funerary practices, dress, ornaments, agricultural techniques, cooking techniques, gaming stones, architecture, copper

Additional artefacts

Picrolite and bone ornaments

Cruciform figurines and figures associated with child-birth, ceramic vessels, picrolite figurines

Cooking pots and baking pans and ovens Abstract ‘plank’ female figures, and genre scenes showing multiple activities

Architecture

Free-standing, monocellular, rectilinear stone and mud brick structures, hearths, pisé and stone benches, partition walls

Single-roomed circular houses Mud-wall construction Central heaths, pits Limited rebuilding and reuse

Multi-roomed rectilinear houses Mould-made mud-brick construction Hearths against side walls Renovation and rebuilding

Burial

No evidence of grave goods. Possible shift to separate dead from living although evidence is too limited

Pit graves within settlements Limited quantity of grave-goods, Bell-shaped shaft burials LChal: cemeteries and picrolite stop

Rock-cut chambers in extramural cemeteries Large quantity of grave-goods

Textile production

No evidence

LChal- crucibles and spindle whorls

Low-whorl spinning Vertical warp-weighted looms Clay weights for warp-weight looms

Ceramic

Coarse ware and Dark burnished ware Household level, open hearths, handmade vessels, hemispherical bowls, ovoid jugs, bottles and holemouth jars

Vessels without handles Painted decoration No direct-fire boiling pots Flasks, bowls, jars, goblets, bottles LChal- new pottery RP jugs and bowls, RW replaced by RB stroke-burnished, changes in firing techniques and shift to specialized production

Vessels with handles Incised decoration Specifically made cooking pots Cooking pots, baking pans, braziers, ovens Direct-fire-boiling and serving vessels, spouted jugs, juglets, small jars, amphorae, and flasks RP Philia

Settlement distribution

Concentration on richer, better watered coastal regions

Concentration on richer, better watered coastal regions

Inland areas of lower rainfall near to copper resources

No evidence

First evidence, chisel and hook, blades, snake ornament from Souskiou

Ore extraction and copper production for tools, weapons and ornaments, spearheads, daggers, axes, razors, tweezers, awls, rings, earrings, and bracelets

Metallurgy

Table 2.3 Summary of differences between Cypriot Ceramic Neolithic, Chalcolithic and Bronze Age cultural phases (data taken from Webb 2001; Horwitz et al. 2004; Frankel 2005, 21; Simmons 2004; McCartney et al. 2006; Guilaine and Briois 2001; Peltenburg et al. 2003; Todd 2003;Le Brun 2001; Clarke et al. 2007; Peltenburg 1978; Clarke et al. 2007; Clarke 2001; Steel 2004;Todd 1991; Peltenburg et al. 1985; Peltenburg et al. 2003; Peltenburg et al. 2006; Peltenburg et al. 1998; Manning 1993; Peltenburg 1996; Frankel et al 1996; Frankel and Webb 2006; Swiny 2008)

23

CHAPTER 3 ARCHAEOBOTANICAL MATERIALS AND METHODS site in 1997 and intensive excavations continued from 2005-2008. It lies on two adjacent modern agricultural terraces at an elevation of ca. 460m above sea level. Notwithstanding damage caused by the construction of the agricultural terraces and natural erosion processes, evidence of Aceramic Neolithic occupation is preserved. The total exposure of the site to date is ca. 288m². Excavations have established the presence of an extensive chipped stone assemblage, a distinctive architecture of circular stone platforms and rubbish pits, and a faunal assemblage that includes cattle (Simmons 2005, 25) (Figures 3.2 and 3.3).

3.1 INTRODUCTION In this chapter there will be a discussion of the archaeobotanical materials analysed in this book and the field and laboratory methods used for their recovery. Presented in the first section is a description of the archaeology of the four sites from which the archaeobotanical samples discussed in this research were recovered. The methods used in the separation of the charred remains from the soil matrix and the laboratory methods used in identification will be presented. This will be followed by a discussion of the methods used for the compilation of the archaeobotanical database assembled from sites located in the Levant, Anatolia, Egypt, and Cyprus and dated to the Aceramic Neolithic, Ceramic Neolithic, Chalcolithic, and Bronze Age. The chapter will conclude with a discussion of the quantification and statistical methods used in this research. 3.2 SAMPLING ON SITE: AN OVERVIEW A total of 8,721 litres from 217 samples were processed for the analyses of charred plant materials. The samples from Krittou Marottou-‘Ais Yiorkis, PrastionMesorotsos, Souskiou-Laona, and Kissonerga-Skalia were processed and analysed between 2005 and 2010. The author could not be involved at all times in the planning of sampling strategies for the four sites due to limited specialist budgets. As a result there are differences between the sites in the quality and quantity of samples. Labour and availability of water were not limiting factors in the determination of the quantity of soil sampled and as a result volumes of soil sampled were appropriate. Soil was sampled from a variety of archaeological contexts including midden and fire pit fills, occupation levels, floors, hearths, and pot spreads. The samples were stored off-site until such time that they could be processed by method of flotation.

Figure 3.1 Krittou Marottou-‘Ais Yiorkis site photographs; top photo was taken in spring and the bottom in summer (the foreground of the bottom photograph shows circular platform feature) (Photographs courtesy of Alan Simmons)

3.3 KRITTOU MAROTTOU-‘AIS YIORKIS ARCHAEOLOGICAL BACKGROUND

The chronology for Cypro-MPPNB occupation is based on 23 radiocarbon dates from animal bone (including cattle), charred macro botanics, and wood charcoal. Of the 23 dates, two come from grains of two-grained einkorn and one from barley. One sample comes from a 160 litre flotation sample from Feature 4 (level 6), and gave a radiocarbon date of 7600-7510 cal. BC (radiocarbon date, Beta-213412 comes from SFN 28 recovered from level 4.6, Feature 4/ unit 20N40W SWQ) (Simmons pers. comm.). The second sample comes from a 150 litre flotation sample from Feature 4 (level 8) and provides a date of 7590-7450 cal. BC (radiocarbon date, Beta-213415 comes from SFN 37 recovered from level 4.8, Feature 4/ unit 20N40W SWQ) (Simmons pers comm.). The radiocarbon dates thus far establish this

Krittou Marottou-‘Ais Yiorkis is a Cypro-Middle to Late PPNB site located in the foothills of the Troödos Mountains (Figure 3.1, photograph of site). The site is located ca. 25 km northeast of the modern town of Paphos, in the upland village of Krittou Marottou. Krittou Marottou-‘Ais Yiorkis overlooks the Ezousas River, which is ca. 1km away, with a spring located ca. 300m south (Simmons 1998, 2). It was first recorded during the Palaeopaphos Survey by Rupp et al. (Rupp et al. 1984, 152) and was thought to be a small upland site related to either deer and/or pig exploitation (Simmons 1998, 2005). Dr Alan Simmons of the University Nevada Las Vegas (UNLV) began test excavations at the upland 24

CHAPTER 3 ARCHAEOBOTANICAL MATERIALS AND METHODS upland occupation within the Middle Cypro-PPNB, c. 7500-7900 cal. BC.

composed of pig (30% of assemblage), caprines (16.9%), cattle (< 2% of the assemblage) and a small amount of cat and dog (Simmons pers. comm.). Of particular significance is the presence of cattle, which were introduced to the island during the Cypro-PPNB, disappeared during the Late Aceramic Neolithic, and then was re-introduced at the beginning of the Bronze Age. FIELD SAMPLING AND CONTEXTS

The author was present to oversee the recovery of archaeobotanical material during all four excavation seasons. A total of 3,084 litres from 42 samples were collected and processed. Appendix 1 is a list of the samples including the relevant sample information, context type, and the volume processed for each sample. Samples were collected from pit fills and occupation levels including the surfaces of two circular platform features. The samples were stored off-site until such time that they could be processed. All samples from Krittou Marottou-‘Ais Yiorkis were processed at either Lemba Archaeological Research Centre (hereafter LARC) or Kouklia Palaeopaphos Museum, where there was water supply available for processing.

Figure 3.2 Photographs of artefacts recovered from Krittou Marottou‘Ais Yiorkis: a) section of Feature 4 showing chipped stone concentration, b) right, stone vessel; left plaster vessel c) picrolite vessel d) chipped stone, e) obsidian artefacts (courtesy of Alan Simmons)

3.4 PRASTION-MESOROTSOS ARCHAEOLOGICAL BACKGROUND

Prastion-Mesorotsos is a multi-period site located 15 km north of Old Paphos in the Dhiarizos River valley. Investigations began in 2009 and the site is currently being excavated by Dr Andrew McCarthy with the University of Edinburgh. Results from the first two seasons (2009 and 2010) revealed a site that covers ca. 10 ha and dates to the following periods: Neolithic; Early, Middle and Late Chalcolithic; possibly Philia, Early, Middle and possibly Late Cypriot I, Archaic/Geometric, Hellenistic, Late Roman, Medieval, and post-Medieval occupation (McCarthy 2010 pers. comm.). Archaeobotanical samples have been taken from several contexts from 2009 - 2011; however, this research presents the results from the 2009 field season and contexts dated to the Neolithic and Chalcolithic only (Areas V and VI). Thus, the following discussion summarises preliminary results from Areas V and VI (McCarthy pers. comm.).

Figure 3.3 Photograph of circular platform, Feature 17 (Photographs courtesy of Alan Simmons)

Krittou Marottou-‘Ais Yiorkis has a series of pits, a possible ditch, and multiple circular platform structures and for which there are no Cypriot or mainland parallels. One of the oval pits, Feature 4, measured ca. three meters in diameter and over one meter in depth. It was excavated between 2004 and 2006 and recovered nearly 10,000 chipped stone artefacts and a large amount of faunal remains. Feature 4 also contained charred plant material (Espinda 2007). Feature 17 is a unique circular platform feature with a plastered surface and plasterlined pit covered with a layer made by mixing chalk and water (Simmons pers. comm.). The material culture includes 42 imported obsidian bladelets (one burin), 16 rare projectile point resembling Byblos types, picrolite ornaments and vessels, carnelian bead fragments, limestone/plaster stone vessels, sickle blades, axes, ground stone for food processing, including hand stones and grinding slabs, and nearly 200,000 pieces of chipped stone. The chipped stone assemblage is typical of the Cypro-PPNB and there are mainland technological and typological parallels. However, there are no similarities with the assemblages of the subsequent Khirokitian. Feature 4 contained an infant burial, possibly in association to the limestone/plaster vessels. The faunal assemblage includes a large percentage of Persian fallow deer (over 50% of the assemblage) with the remaining

Areas V and VI are situated on the lower terrace south of a rocky outcrop, both have been affected by natural erosion processes (for site plan refer to Figure 3.4). Area VI is situated on the terrace above Area V. Chronology of the site has so far been based on material culture (principally ceramic and lithic assemblages); with Area V dated to the Aceramic Neolithic and Area VI dated to the Late Ceramic Neolithic/Early Chalcolithic transition. Excavated features include rubbish pits, walls, and ephemeral structures (Features 544, 545, and 546) (Figure 3.5). Context 566 (fill of rubbish pit 556) is the earliest occupation level excavated so far and is likely to 25

CROPS, CULTURE, AND CONTACT date to the Cypro-PPNB. Preliminary chipped stone analyses show a Late Neolithic industry with technological parallels with the southern Levant. The ground stone assemblage consists of plant processing tools, including querns, rubbers, grinders, pestles, bowls, mortars, and pounders. Preliminary analyses of the faunal assemblage include pig, caprines, deer, fox, and claws of freshwater crab, fish, and bird (McCarthy 2010 pers. comm.).

FIELD SAMPLING AND CONTEXTS

The author was involved in the processing of a small portion of the archaeobotanical material from PrastionMesorotsos (2009) but was not involved in on-site sampling and sampling strategies. A total of 980 litres were processed from 19 samples collected in 2009 from Areas V and VI. All were processed at either LARC or Kouklia Palaeopaphos Museum, where there was water supply available for processing. The samples were collected from midden pit fills and occupation levels. Appendix 2 is a list of the samples from Areas V and VI including the relevant sample information, context type, and the volume processed for each sample. 3.5 SOUSKIOU-LAONA ARCHAEOLOGICAL BACKGROUND

Souskiou-Laona is an early Middle/Middle Chalcolithic settlement site located ca. 2.5 km inland from the modern village of Kouklia on the island’s southwest coast. The site measures ca. 2.2 ha. and is located ca. 300m southwest of the Souskiou Chalcolithic cemetery complex (Souskiou-Laona and Souskiou Vathyrkakas 13, Peltenburg et al. 2006). The settlement is currently being excavated by Professor Edgar Peltenburg with LARC and the University of Edinburgh. Excavations began in 2005 following the completion of excavations of the cemetery complex (Peltenburg et al. 2006). The following summary is based on information from unpublished season reports and from Peltenburg et al. (2006).

Figure 3.4 Area V Plan of Prastion-Mesorotsos showing location of excavation Areas/Trenches (1-8). Plan courtesy of Andrew McCarthy)

Souskiou-Laona settlement is located on top of a level ridge that drops steeply on three sides which has caused the damage to the site as a result of natural erosion processes. Views from the ridge include the Troödos Mountains to the Northeast, the Mediterranean Sea to the south and the Dhiarizos and Vathyrkakas rivers valleys to the east and west (Peltenburg et al. 2006) (Figure 3.6). There are three operations/areas of the site that have been excavated: Operations A, B, and D (Figure 3.7). Operation A is the best preserved area and is located on the lower slope of the East Ridge, Operation B is located on the Northeast Ridge and Operation D is located on the West Ridge. Architecture includes multiple characteristic Middle Chalcolithic circular structures, hearths, fire installations, and rubbish pits (composed of midden-fill and slope-wash) (Figure 3.8). Also, there is evidence of human remains (three children and one adult) in pitgraves located within Building 648 (Operation A, Trench 1).

Figure 3.5 Photograph of features at Prastion-Mesorotsos; a) Area V, b) Area VI, c) Rubbish Pit Feature 501(Photographs courtesy of Andrew McCarthy)

Chronology for the site has so far been established based on ceramic typologies, with three chronological phases: Chalcolithic, early Middle Chalcolithic, and late Middle Chalcolithic. An absence of Early Chalcolithic Glossy Burnished Ware suggests an early Middle Chalcolithic date for initial occupation (Peltenburg et al. 2006). Pottery excavated from Operation A is dated to the Middle Chalcolithic, including Red Monochrome

Figure 3.6 Photograph of Souskiou-Laona (top photograph was taken in spring; bottom photograph was taken in summer) (courtesy of E. Peltenburg)

26

CHAPTER 3 ARCHAEOBOTANICAL MATERIALS AND METHODS Painted Ware and Red-on-White Parallel Band ware. Operation B is dated later than Operation A, with pottery types assigned to late Middle Chalcolithic (i.e. RWL, RMP-b, SE, and CPW-mono). The ceramic types include platters, hemi-bowls, deep bowls, spouted bowls, trays, flasks, goblets on stands, and storage jars. However, the storage jars are rare, which is typical of Middle Chalcolithic pottery assemblages (i.e. KissonergaMosphilia and Kissonerga-Mylouthkia) (Peltenburg et al. 2006).

ornaments. The ground stone assemblage includes adzes, axes, hammerstones/grinders, pestles, rubbers, chisels, and querns. Additional material culture includes chipped stone, bone and antler objects, terracotta, and picrolite wasters and figurines. FIELD SAMPLING AND CONTEXTS

The author was not able to be present to oversee on-site sampling and recovery of the samples taken from Souskiou-Laona, with the exception of the 2010 season when it was possible for the author to conduct flotation. A total of 2,137 litres from 64 samples were collected and processed from the 2004-2010 seasons. The results from all seasons are included in this book with the exception of samples collected from the 2009 field season. Samples were collected from building floors, occupation levels, and midden pit fills. Appendix 3 is a list of the samples including the relevant sample information, context type, and the volume processed for each sample. 3.6 KISSONERGA-SKALIA ARCHAEOLOGICAL BACKGROUND

Kissonerga-Skalia is an Early/Middle Bronze Age (ECMC, ca. 2400-1650 BC) settlement located 300 m from the southwest coast in the modern village of Kissonerga in the Ktima lowland just south of Kissonerga-Mosphilia (c. 6000-2400 BC). The site has been excavated since 2007 under the direction of Dr Lindy Crewe with the University of Manchester. The following summary is from unpublished field season reports (Crewe & Hill 2012).

Figure 3.7 Photograph of circular foundation Souskiou-Laona (courtesy of E. Peltenburg)

Notwithstanding damage caused by machine terracing and ploughing for agricultural activity in the 1970s and agricultural activity dated to the Medieval period, the results thus far establish Early and Middle Bronze Age occupation; a period that has not yet been investigated in this region. As yet there are no 14C dates but relative chronologies have been established on the basis of pottery. The pottery includes Drab Polished Ware, Red Polished Ware, late White Painted V-VI sherds, Black Slip handmade, and Plain White handmade pithoi. Excavation has not reached a sufficient depth to make a conclusion regarding initial date of site occupation. However, Crewe suggests, at this time, it is possible that the site could have been occupied at the beginning of the Early Bronze Age and that the occupants could have relocated south from Kissonerga-Mosphilia. Thus, results from the site have the potential to contribute to knowledge of the cultural transition between the Chalcolithic, Philia, and Early Bronze Age on the island’s southwest coast.

Figure 3.8 Souskiou-Laona site plan of the location of Operations, trenches, and settlement in relation to Souskiou-Laona cemetery (courtesy of E. Peltenburg)

The site differs from contemporary sites with regards to the distribution and possible significance of the material culture. The total number of picrolite (ornaments and wasters) and metal ornaments (i.e. copper) found within the structures at, or around, the time of abandonment have raised questions regarding the function of the site in a regional context (Peltenburg et al. 2006). The ornamental artifacts associated with birthing and death in correlation with the abandonment of the structures has lead Peltenburg et al. (2006, 85) to view the settlement as a possible regional centre for the distribution of symbolic material culture. For instance, in Operation D (Trench 30) there is a building that has been interpreted as a picrolite sculptor’s workplace due to its large number of picrolite wasters and dentalium body

Excavations are now concentrated on the upper agricultural terrace (Plot 199) (for site plan refer to Figures 3.9 and 3.10). Trenches B, D, and G-G2 have so far revealed interesting architectural and cultural finds. 27

CROPS, CULTURE, AND CONTACT Trench B was excavated between the 2007-2010 seasons and the architecture includes a large curvilinear building (Feature 33) with an associated plaster-floored courtyard, a large furnace-like structure framed by two stone wall foundations and a series of fire pits, some of which are pottery- or stone-lined (by partial pottery vessels and sherds) (Figure 3.11). One particular fire pit was lined with a large Red Polished ware pithos jar. Trench D is earlier than Trench B, with architecture more typical of the Cypriot Early/Middle Bronze Age with a multiroomed rectilinear structure. Also included in Trench D are pits, a stone and mud-plaster bin, a hearth/fire pit and multiple cooking pots and storage vessels. Trench G - G2 has plaster- and pithoi-lined pits, pots spreads, groundstone tools, and a crudely constructed large wall. Cultural materials recovered from excavations thus far include beads and pendants, copper fragments, spindle whorls, a loom weight, and multiple querns and gaming stones. Preliminary results from faunal data include remains of cattle, deer, pig and sheep/goat, crab and shellfish.

Figure 3.10 Kissonerga-Skalia site plan of Plot 199, excavation trenches, and outline of a selection of archaeological features (courtesy of L. Crewe pers. comm. 2012) FIELD SAMPLING AND CONTEXTS

With the exception of the 2009 field season, the author was involved in on-site sampling strategies and recovery of samples. A total of 2,519 litres from 92 samples were collected and processed from the 2007-2010 excavation seasons. Samples were recovered from hearths, plaster floors, occupation levels, pot spreads, rubbish and fire pit fills and from the fill of a mud plaster ‘furnace’. All samples were processed at LARC. Appendix 4 is a list of the samples including the relevant sample information, context type, and the volume processed for each sample.

Figure 3.9 Kissonerga-Skalia site plan showing the excavation areas (courtesy of L. Crewe pers. comm. 2011)

3.11 Kissonerga-Skalia, photograph of Trench B, Plot 199 (courtesy of L. Crewe pers. comm. 2012)

28

CHAPTER 3 ARCHAEOBOTANICAL MATERIALS AND METHODS 3.7 RETRIEVAL OF PLANT MATERIAL

3.9 IDENTIFICATION

The macro botanical remains from the four sites were preserved in charred form and were separated from the soil by systematic water flotation. The methods used in the recovery and processing of the data analyzed here are comparable to methods used previously in Cyprus including the use of a low pressure water flow tap for flotation; sieve sizes of 1 mm and 250 μm; and the use of a low power binocular microscope for identification (Colledge 2003; Murray 1998; 2003). When the author was not there to oversee recovery of charred plant material, multiple students, including undergraduates from the University of Edinburgh and University of Manchester field schools, were able to process the samples.

The charred material was sorted and all identifiable plant macro-remains (e.g. seeds, nuts, chaff, and charred wood charcoal) were extracted from the flots and analysed under a low power binocular microscope. Wood charcoal was separated from the charred plant remains and sent to Katleen Deckers at Universität Tübingen for analysis. Identifications of the charred material were made by comparing taxa with specimens in the modern reference collection of plant taxa (i.e. seeds, fruits, nuts, etc.) housed at the Institute of Archaeology, University College London, which comprises a majority of accessions that were collected by Professor Gordon Hillman in Turkey, Syria and Jordan. Photographs, drawings, and descriptions of plant taxa were also used to aid identification and with reference to the following publications: Zohary and Hopf (2000), Jacomet (2006), Nesbitt (2006), and van Zeist and Bakker-Heeres (1982, 1984, and 1985). It was impossible to identity all plant remains to species level and thus some have been identified to genus level (e.g. Triticum) or to the family level (e.g. Leguminosae) only. The abbreviation ‘cf.’ is used when a specimen compares with or most closely resembles a particular species or genus. Identification criteria for the taxa identified from the four sites analysed are presented in Appendix 5 and photographs of charred plant specimens are presented in Appendix 6.

Due to lack of water on-site, the samples were processed at LARC and the museum at the Kouklia Palaeopaphos Museum where there was running water available for flotation, either from a tap or from a spring. The excavated deposits were not screened prior to flotation. The smallest mesh size used to retain the charred plant remains was a 250 μm mesh ‘flot bag’ or two metal sieves, with mesh sizes of 1mm and 250 μm. Samples were floated in an eighty-five litre metal barrel and the tank was cleaned after each sample had been processed (e.g. when the fine sediment had accumulated to a considerable depth in the bottom of the tank) to avoid cross-sample contamination. Within the barrel was a 1 mm mesh used to catch heavy fractions. The heavy fractions were labeled and dried out of direct sunlight and thereafter sorted for small artefacts and any dense non-floating plant remains at LARC and/or Kouklia Palaeopaphos Museum.

3.10 COMPILATION OF DATABASE All previously published archaeobotanical material from Cyprus, data from contemporary mainland sites, and the new botanical data presented here have been entered into a relational database (Microsoft Access) (cf. the database design described in Colledge et al. 2004). The decision to enter the data in a relational database similar to the design described by Colledge et al. (2004) is because it facilitates comparative analysis between datasets. The data compiled here include the Cypriot database compiled by the author and two separate mainland databases compiled by Colledge et al. (2004) and Simone Riehl. The three databases were amalgamated by Sue Colledge.

3.8 SORTING OF THE LIGHT FRACTION The light fraction was exported with the permission of the Department of Antiquities of Cyprus to the University College London for analysis. The flots (light fraction) were further separated into two fractions: 1mm (fine flot). These fractions were then sorted separately to make the process of identification easier as the eyes accommodate and recognize shapes of the same size more efficiently. Both fractions were sorted and the results thereafter combined. All of the coarse flot was sorted and identified; however, not all of the fine flot was sorted. The decision to sort only a small portion of the fine flot, a quarter of the sample in some instances, was based on time constraints. Fine flot is often more time consuming because of the tiny, and often numerous, charred items that have to be extracted. If only a subset of the fine fraction had been sorted it was necessary to ‘multiply up’ the total of number of items to represent 100% (e.g. if only ¼ sorted the totals were multiplied by 4) prior to calculating the numbers of taxa for both. If the coarse flot did not contain any charred plant material (e.g., including wood charcoal) a decision was made not to continue to sort the fine flot.

3.11 QUANTIFICATION OF THE REMAINS COUNTING TAXA The total number of items (e.g. seeds, fruit stones, nuts and chaff elements) and ‘whole item equivalents’ are counted in the list of taxa. For cereals, the whole grain equivalent was calculated by counting the number of either apical or embryo fragments of wheat, barley and grasses, whichever was the largest, and the larger of the two was the total number of whole grains. Indeterminate cereal grain fragments were converted to ‘whole grain equivalents’ on the basis of weight. Krittou Marottou‘Ais Yiorkis was the only site that had enough whole cereal grains to calculate a ‘whole grain equivalent’ for cereal grain fragments. For Krittou Marottou-‘Ais 29

CROPS, CULTURE, AND CONTACT Yiorkis, five ancient charred grains were weighed, three whole grains of T. monococcum 2g and two whole grains of H. sativum and the average weight of one grain was calculated to be 0.05 grams. However, Colledge (1996, 66) calculates the whole grain equivalent for cereal grain fragments as one grain equal to 0.009 grams. The whole grain equivalent calculated for Krittou Marottou-‘Ais Yiorkis is large in comparison, so the decision was made to calculate the whole grain equivalent for the other sites based on calculations used by Colledge (1996, 66). The weights of the cereal fragments were divided by the average weight of one grain to calculate the numbers of ‘whole grain equivalents’. Pistacia sp. nutshell fragments were converted to whole nut equivalents on the basis of weight following the calculation by Colledge (1996, 66). Colledge (1996) calculated the weight of three whole nuts to be 0.07 g. Whole pip equivalents for Vitis sp. was calculated as four fragments equal to one whole pip. Whole seed equivalents for legumes were calculated based on the number of halves; two cotyledons equal to one whole seed and four fragments equal to one whole seed.

questions of this research includes many plant taxa (variables) and multiple archaeological sites from several cultural phases (samples). Multivariate techniques are useful for comparisons of material from different sites because they help reduce noise as a result of variations in preservation, sampling, recovery, and identification. Lange (1990) states, with regards to multivariate techniques: “Redundancy of information is summarized, noise is reduced, outliers can be identified and relations brought to light.” CORRESPONDENCE ANALYSIS Correspondence analysis (hereafter CA) is a multivariate statistical technique that is useful in the analyses of abundance data and thus is the method that will be used here to simplify the complex data set. CA graphically displays the relationships between complex datasets (Bølviken 1982). Lange (1990, 43) states, “In graphical form the results of a Correspondence Analysis bring out the position of each sample relative to all other samples and to all the species, and of each species relative to all other species and to all the samples in the analysis” (Lange 1990, 43). Further, “with correspondence analysis the relationships between cases, those between variables, and those between variables and cases, may all be analysed together and represented in the same scattergram or series of scattergrams” (Shennan 1988, 284). The aim by using CA is to demonstrate any temporal patterning in the samples/sites in the dataset that is associated with specific plant use/exploitation. The computer software used to perform CA was CANOCO (Ter Braak 1988). CANODRAW (Smilauer 1992) was used to graphically plot the output from the analyses.

3.12 STATISTICAL METHODS UNIVARIATE METHODS The data from Krittou Marottou-‘Ais Yiorkis, PrastionMesorotsos, Souskiou-Laona, Kissonerga-Skalia, and the previously published material from the mainland Levant and Cyprus will be described on the bases of analysis of presence, density, diversity (e.g. by calculating diversity indices) and ubiquity (Wright 2010; Jones 1991). The average number of cereals per litre in each sample was used as proxy for the measure of relative density of charred remains. Percentage presence, or ubiquity, for each taxon is the percentage of the number of samples in which the taxon occurs. Since it is unlikely that absolute numbers reflect original proportions or importance in the past, percentage presence is used (Jones 1991). Ubiquity is also useful in comparisons of data that are the result of various taphonomic and retrieval processes and different identification and recording styles. Also, presence/absence will be used in comparisons between archaeological sites from the mainland Levant and Cyprus since for some sites presence/absence data is all that has been recorded and/or published. Presence/absence has been shown to be a useful level of analysis in comparative investigations of large datasets. Further, the information that is eliminated at the level of presence/absence (i.e. absolute counts) has been shown to influence the larger trends in datasets only minimally, thus the important trends in the data are shown at the level of presence/absence (Lange 1990; see also discussion in Colledge et al. 2004). MULTIVARIATE METHODS Multivariate analysis is useful for archaeobotanical datasets that include multiple variables and many samples. The dataset compiled to address the research 30

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY 4.1 INTRODUCTION

4.4 INTRODUCTION OF FLOTATION TECHNIQUES IN CYPRUS

In this chapter there will be a brief discussion of the history of the use of flotation for the recovery of plant remains in Cyprus. Different methods of sampling, recovery and identification have been used on sites in Cyprus and a summary of these will be presented, in addition, previous archaeobotanical research undertaken on Aceramic Neolithic to Late Bronze Age assemblages will be described.

The introduction of flotation techniques is significant in the history of archaeobotany worldwide. Flotation facilitated much larger quantities of plant remains (comprising both large and small taxa, as opposed to hand-picking) to be recovered that were much more likely to be representative of the full suite of charred remains preserved in occupation deposits. Standardisation of recovery systems thus meant that comparison of archaeobotanical data between periods, sites and regions was possible (i.e., not limited due to the likelihood of differential recovery of plant taxa). The first experiments with flotation began in 1962 on the Lowillva project in Illinois, and in 1968 Streuver published a description of the process for which charred material could be separated from excavated soils sampled from archaeological sites (Streuver 1968). In 1963 Hans Helbaek at Deh Luran in Iran modified the technique by using a bucket flotation system, which utilized a ‘wash-over’ technique as opposed to a ‘scoop’ method (Hole et al. 1969; Helbaek 1969). In 1970 van Zeist published the results of material floated at Mureybet, Syria (Van Zeist 1970) and in 1972 Jarman et al. (1972) published a paper on the recovery of plant remains by froth flotation with trial flotation at the Bronze Age to Medieval site of The Udal in North Uist (Outer Hebrides, Scotland) and at Early Neolithic Nahal Oren in Palestine. In sum, experiments with flotation techniques began in the early 1960s and the application of flotation was rapidly adopted in the old and new worlds, and had considerable implications for all aspects of archaeobotanical research.

4.2 PRESENTATION OF TAXA Table 4.1 is a list of Cypriot archaeobotanical publications that include site-based reports and summaries, and syntheses covering sites dated to the earliest phases of island occupation to the Late Bronze Age. The recovery methods used, the number and volume of samples, and whether any information regarding provenience of the plant material for each site dated from the Aceramic Neolithic to the Middle Bronze Age is presented in Table 4.2. Presence/absence records for cereals and non-cereal taxa from sites dated to the Aceramic Neolithic to the Middle Bronze Age are listed in Appendix 7. The taxa recovered from sites dated to the Late Bronze and later have been combined for a more general comparative discussion in Appendix 7. The cereal taxa are presented first followed by trees/shrubs and wild herbaceous taxa listed in the order and nomenclature of the Flora of Cyprus (Meikle 1977, 1985). The total number of taxa presented in the bar charts discussed in sections 4.3-4.8 below omits presence at the family level. For a list of crop and tree species that first appear in each cultural phase refer to Table 4.3.

The archaeobotany of prehistoric Cyprus has its origins before the introduction of modern flotation techniques in the 1960s. In 1952, botanical results from Late Bronze Age Apliki-Karamallos were published. The charred plant material was hand-picked at the time of excavation and sent to the Department of Agriculture in Nicosia for analysis. The results included a list of plant taxa (du Plat Taylor 1952), and decades later charred specimens were used for radiocarbon dating (Kling et al. 2007). Additional material that was hand-picked during excavation at Apliki-Karamallos and not previously examined was analysed by Helbaek nearly a decade later (Helbaek 1962). Helbaek concluded from the results that there was exploitation and cultivation of locally available plant resources including bread wheat, six-row barley, lentil, and horse bean (Helbaek 1962); he was the first to interpret Cypriot prehistoric botanical data. It was Stewart (1974) who first applied modern flotation techniques in Cyprus at Dhali-Agridhi. This was followed by flotation at Dhali-Agridhi, KhirokitiaVounoi (Waines and Stanley-Price 1977) and Cape Andreas-Kastros (van Zeist 1981).

4.3 CHRONOLOGY Radiocarbon dates have been re-calibrated using Ox Cal v3.10, with the IntCal09 calibration curve (Bronk Ramsey 2006; Reimer et al. 2004). All chronological determinations are expressed in calibrated years BC. Unless otherwise stated, the average calibrated date is based on a point estimate age with 1-sigma of the summed probability (one standard deviation), which will be used in this discussion to assess an estimated chronology for crop and weed introductions to the island. The calibrated radiocarbon dates are presented in Appendix 8 and the summed average calibrated dates are illustrated in Figure 4.1 (with references). The dates in Figure 4.1 represent the calibrated dates from sites with archaeobotanical data, and the dates illustrate relatively continuous occupation from the Aceramic Neolithic (ca. 8500 cal. BC) to the Middle Bronze Age (ca. 1900 cal. BC), the time period this research examines.

31

CROPS, CULTURE, AND CONTACT

Figure 4.1 Calibrated radiocarbon dates based on one standard deviation. Kissonerga-Mylouthkia (n=5 ) (Peltenburg 2003); PerekklishaShillourokambos (n=9) (Guilaine 2003); Kalavasos-Tenta (n=16) (Todd 2005); Dhali-Agridhi (n=5); Cape Andreas-Kastros (n=3) (le Brun 1981, p. 71); Kholetria-Ortos (n=6) (Simmons 1994); Khirokitia-Vounoi (n=17) (Le Brun 1994, 1991); Ayios Epiktitos-Vrysi (n=17) (Peltenburg 1982c); Kantou (n=2); Lemba-Lakkous (n=9) (Peltenburg 1985)); Kissonerga-Mosphilia (n= 30) (Peltenburg 1998); Chalcolithic Kissonerga-Mylouthkia (n=9) (Peltenburg 2003); Kalavasos-Ayious (n=4) (Todd and Croft 2004); Marki-Alonia (n=9) (Frankel and Webb 1992); Sotira-Kaminoudhia (n=9) (Swiny et al.2003)

Figure 4.2 Cumulative number of publications and number of sites in Cypriot archaeobotany with time, The darker line denotes number of publications and the lighter line indicates number of sites that have been added to the total number of sites with archaeobotanical data

32

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY remains were recovered by flotation, using both 1mm and 250 micron mesh sieves (Murray 2003, 59). By using both mesh sizes the chances of recovering very small seeded species increase. Identification was conducted with the aid of a low power microscope (Murray 2003), which helps to identify differences in morphologies that exist between genera and species. Adams and Simmons (in Frankel et al. 1996) give no information regarding sieve size, only that a combination of flotation and wet sieving was used at Marki-Alonia for the recovery of plant material, which was sorted without the aid of a microscope (in Frankel et al. 1996, 224). Taxa identified at Kissonerga-Mylouthkia (Colledge 2003; Murray 2003) and Marki-Alonia (Frankel et al. 1996) are likely the result of different retrieval and identification methods as well as different post-harvest processing stages. Table 4.2 lists the recorded number of samples, volume of samples, preservation and recovery method (including details of sieve sizes used), and whether whole counts or presence only records are given and contextual information for sites discussed below.

4.5 CURRENT ARCHAEOBOTANY IN CYPRUS To date there are 50 publications on the archaeobotany of Cyprus. These include site-based reports and summaries and syntheses covering sites dated to the earliest phases of island occupation to the Late Bronze Age (Table 4.1). The number of publications increases over time with one publication in the 1950s, two in the 1960s, 13 in the 1970s, ten in the 1980s, eight in the 1990s, and 16 since 2000 (Figure 4.2). There are a total of 30 sites (some with data from multiple cultural phases) in Cyprus that have archaeobotanical data, this includes the four sites analysed in this research. Archaeobotanical results from Aceramic Neolithic and Late Bronze Age periods are well documented but there are far fewer reports from the Ceramic Neolithic, Chalcolithic, and Early to Middle Bronze Age. There are seven sites with archaeobotanical data from the Aceramic Neolithic, three from the Ceramic Neolithic, six from the Chalcolithic, four from the Early/Middle Bronze Age, and eight from the Late Cypriot period and later. In addition to the disparity in cultural phase representation there are differences in the representation of botanical data. The ‘unevenness’ in the quantity of plant specimens and the number of taxa recovered from sites has been previously highlighted (see Zohary and Hopf 2000, 247) and further discussed in studies in the origins and spread of crop-based agriculture (Colledge et al. 2004, S44). Many archaeologists and archaeobotanists have been responsible for implementing sampling for the recovery of archaeobotanical plant remains recorded from sites dated to the prehistoric phases of Cyprus. Thus, there is great diversity in sample size and numbers, equally varied is the care with which records have been kept since the introduction of more efficient recovery methods. Moreover, as is common elsewhere in the Near East and eastern Mediterranean, on Cyprus there is not, nor has there been in the past, a standard methodology for the recovery of plant materials. As a result variations in excavation, sampling and archaeobotanical processing techniques have certainly contributed to differences in the taxa represented in botanical assemblages. In particular, differences in archaeobotanical recovery techniques such as hand-picked versus flotation methods, sieve size used in flotation, the number of litres per sample, and whether a microscope was used to aid in processing has directly affected the number of taxa, the quantity of charred remains, and species representation. For example, Stewart (1974) records a total of 12,090 litres sampled at Dhali-Agridhi and Adams and Simmonds (in Frankel et al. 1996, 224) record 235 total litres sampled from Marki-Alonia. Sample sizes have varied greatly in other regions as well, Willcox et al. (2008, 315) report that 12,114 litres were processed at Jerf el Ahmar, 1,520 litres at Tell ‘Abr, 6,122 at Dja’de, and 1,772 litres at Tell Qaramel. Methods of recovery, processing, and identification have also varied. Murray (2003) describes the methodology used in the analyses of material from Aceramic Neolithic Kissonerga-Mylouthkia; the charred

4.6 ACERAMIC NEOLITHIC Seven flotation samples were collected from Epipalaeolithic Akrotiri-Aetokremnos; however, with the exception of small amounts of Pinus sp., Genista-type wood charcoal specimens no other charred material was recovered (Simmons 1999). Likewise, flotation samples from Cypro-PPNA Ayia Varvara-Asprokremmos are currently under analysis by the author but so far no charred material has been recovered. So, the earliest period with recorded archaeobotanical evidence is the Cypro-PPNB. Prior to this study, there were seven sites with botanical evidence for the Aceramic Neolithic of Cyprus. These sites range from the Early Cypro-PPNB to the late Pre-Pottery Neolithic/Khirokitian with a date range from about 8,500 cal. BC to 5,500 cal. BC (refer to Figure 4.1 for chronological ordering of sites). The seven sites below are discussed in decreasing order of age. The sites are Kissonerga-Mylouthkia, ParekklishaShillourokambos, Kalavasos-Tenta, Dhali-Agridhi, Cape Andreas-Kastros, Kholetria-Ortos, and KhirokitiaVounoi . All seven sites have charred botanical remains that were recovered from flotation. The exception is Parekklisha-Shillourokambos which is mostly impressions in pisé with minimal charred plant remains recovered from flotation. All authors recorded the number of samples and the total volume samples, except Dhali-Agridhi, which the volume was not reported. The total number of samples recorded for the Aceramic Neolithic of Cyprus is 877 and the total number of litres sampled is approximately 28,208. KISSONERGA-MYLOUTHKIA Kissonerga-Mylouthkia is located on the southwest coast of the island. The site has two cultural phases, Aceramic Neolithic and Chalcolithic. A total of five radiocarbon dates show two phases of Aceramic Neolithic occupation, Phase 1A and 1B (Peltenburg 2003). Phase 33

CROPS, CULTURE, AND CONTACT 1A dates to ca. 8450 cal. BC and Phase 1B dates to ca. 7150 cal. BC, with a gap of approximately a millennium between. In 2003, Murray and Colledge published the archaeobotanical results from charred remains from the 1976-1996 excavations (Colledge 2003, 239-245; Murray 2003, 59-71). Murray presents the results from the Cypro-PPNB occupation (Colledge presents the results from the Chalcolithic occupation, the latter discussed in the Chalcolithic section below).

PAREKKLISHA-SHILLOUROKAMBOS Parekklisha-Shillourokambos is a ca. 1ha. site located 6km east of Limassol. Nine radiocarbon dates have identified two phases of Aceramic Neolithic occupation at, Phase A (Cypro-EPPNB) and Phase B (CyproM/LPPNB) (Guilaine 2003), with a date range between c. 8425 and 7875 calibrated BC, and an average date of ca. 8075 cal. BC (based on one standard deviation). Architectural remains include pits, wells, circular stone structures, hearths, post-hole alignments, and a trapezoidal enclosure bounded by trenches (Guilaine 2003). Willcox (2001) published the archaeobotanical results from Parekklisha-Shillourokambos. A total of 19 samples (2,446 litres of soil) were floated using a mesh size of 0.5 mm. However, possibly due to extremely poor preservation results are based mainly on plant impressions in pisé and only nine taxa are recorded and (Willcox 2001, 129), including wild barley grain and chaff, and indeterminate glume wheat grains and chaff and indeterminate barley grains, Capparis spinosa (hereafter caper), Prunus sp., and the following wild herbaceous taxa: fumitory, Lathyrus sp., and Galium sp. (hereafter bedstraw). The results from ParekklishaShillourokambos provide additional evidence for colonisation by famers to the island in the Cypro-PPNB. However, the archaeobotanical evidence includes wild barley grain and chaff. This has fuelled debates on whether the plant remains were a result of local agricultural developments or introduced to the island as part of a farming “package”. However, due to the paucity of plant specimens recovered from ParekklishaShillourokambos it is difficult to understand the cause of differences in data at this time (Colledge and Conolly 2007, 59).

The botanical material was preserved in charred form and recovered by flotation using both 1 mm and 250 micron mesh sieves. A total of 880 litres from 12 samples was sampled from two wells, a pit and a building fill. Five samples are from Period 1A and seven samples are from Period 1B. The botanical material includes domesticated cereal grain and chaff, legumes, wild herbaceous taxa (interpreted as potential crop weeds), fruit and oil plants, and nuts (Murray 2003, 5971). For Phase 1A there are four domestic taxa, one oil plant, one fruit, and seven wild herbaceous taxa. The domestic taxa include Hordeum sativum (hereafter hulled barley) grains and chaff, Triticum dicoccum (hereafter emmer wheat) and Triticum monococcum (hereafter einkorn wheat) grains and chaff and Lens sp. (hereafter lentil). There is also evidence of Linum sp. (hereafter linseed), Pistacia sp. (hereafter pistachio), and seven wild herbaceous taxa. In comparison with the earlier samples, Phase 1B has one additional fruit tree, Ficus sp. (hereafter fig) and the following additional wild herbaceous taxa: Adonis sp. (hereafter pheasant’s eye), Fumaria sp. (hereafter fumitory), Malva sp. (hereafter mallow), Scorpiurus spp. (hereafter prickly caterpillar), Rumex sp. (hereafter dockweed), Polygonum sp. (hereafter knotweed), Hordeum sp. (hereafter wild barley), and Beta sp. (hereafter beet).

KALAVASOS-TENTA The botanical data from the Aceramic Neolithic, particularly, the data recovered from KissonergaMylouthkia 1A contributed to debates on the island’s early economic development. The plant assemblage recovered from the earlier phase of the site provides evidence for a colonisation by farmers to the island at ca. 8500 cal. BC and against the argument of agricultural development by local foragers (Peltenburg et al. 2000). Further, the data provides evidence for agricultural continuity with the subsequent Khirokitian culture (Peltenburg et al. 2000), a culture that only in the 1990’s was thought to have been the earliest evidence of human occupation on Cyprus. Evidence of domesticated cereals and their associated weed taxa in early Cypriot Neolithic assemblages also sheds light on the timing and direction of Near Eastern cereal crop dispersal with Cyprus being the first targeted region colonized in the early spread of Near Eastern agriculture (followed by central Anatolia and then Crete and Greece) (Colledge 2004; Colledge et al. 2004).

Kalavasos-Tenta is located 4 km south of Kalavasos in the Larnaca district on the island’s southern coast. The site was excavated over five seasons between 1976 and 1984. A total of 19 radiocarbon dates show two phases of occupation, Aceramic Neolithic and Ceramic Neolithic (Hansen 2005, 178). Five separate phases have been assigned to the Aceramic Neolithic. The earliest phase is Period 5, which is broadly contemporary with the early levels of Parekklisha-Shillourokambos. Period 5 lacks substantial architecture and has stake holes and pits. Period 4 dates to ca. 7500 cal. BC and the architecture includes a site enclosure wall and ditch. Period 3 has a mud-brick building and dates to about 6950 cal. BC. The best represented phase is Period 2 and it is defined by tightly clustered curvilinear stone and mud-brick domestic structures dating to ca. 6200 cal. BC (Todd 2004). Although five phases of occupation have been assigned culturally, the calibrated radiocarbon dates range between ca. 7,600 and 5,800 cal. BC and illustrate three main phases (illustrated in Figure in 4.1 as KTenta I, KTenta II, and KTenta III). For this discussion botanical data from Phase 4 will be represented by

34

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY KTenta I, Phase 3 will be represented by KTenta II, and Phase 2 will be represented by KTenta III.

from the 1972 field season. A total of 12,090 litres of soil was floated from 109 contexts, many of which contained no identifiable plant remains. There are 19 taxa from the Aceramic Neolithic occupation and two from the Ceramic Neolithic occupation (Stewart 1974, 123-129). The domesticated crops from Dhali-Agridhi include both naked and hulled barley, emmer wheat, free-threshing wheat, Cicer arietinum (hereafter chickpea), lentil, Pisum sativum (hereafter pea) and Vicia sp. (vetch). In addition to fig and pistachio there are trees that have not previously been recorded including, Celtis sp. (hereafter hackberry), Olea sp. (hereafter olive), and Vitis sp. (hereafter grapevine), and Prunus domestica (hereafter plum).

In 2005 Hansen published the archaeobotanical results (Hansen 2005, 323-343). The material was preserved in charred form and recovered by flotation. A total of 416 samples were floated totalling approximately 7,764 litres, of which only 175 samples (2,074 litres) had identifiable remains. Samples were taken from both pits and hearths, and inside and outside of domestic structures. Hansen divides the Aceramic Neolithic samples into the following cultural phases, from latest to earliest: Periods 1 – 5. The latest and earliest phases of occupation only have specimens from the grass family and therefore, Periods 1 and 5 have been excluded from this discussion. The data from the Late Neolithic/Chalcolithic transition will be discussed below.

CAPE ANDREAS-KASTROS Cape Andreas-Kastros is a late Aceramic Neolithic site located at the northern most tip of the Karpas peninsula. The site was excavated for four consecutive seasons from 1970-1973. Three radiocarbon dates (le Brun 1981, 71) give an average date of ca. 5850 cal. BC. In 1981, van Zeist published the botanical results from the 1973 field season. The plant remains recovered were preserved in charred form and twenty-three samples were processed using flotation. No contextual information or information on samples size is provided. The quantities are provided for each of the 21 taxa present, which includes hulled barley, glume wheat grains and chaff (emmer and einkorn wheat), lentil, pea, vetch, and Vicia faba (hereafter faba bean). Noteworthy is the presence of faba bean, which has not been recorded in Cypriot samples dated earlier (van Zeist 1981). Van Zeist discusses the likelihood of the presence of six-row barley but due to a high degree of fragmentation and poor preservation the assignment to species could not be made with certainty (1981, 97). However, six-row barley is possible because it is also present in the assemblages of Kalavasos-Tenta. Interesting is the presence of Lolium perenne/rigidum (hereafter ryegrass), which is a weed commonly associated with cereal cultivation. However, due to its quantity (480 grains) and ubiquity (86.9 %), van Zeist suggests the possibility of ryegrass being used as an economic resource.

Plant remains from the Aceramic Neolithic occupation of Kalavasos-Tenta include three domesticated cereals, einkorn wheat (both one and two-grained) grains and chaff, emmer wheat grains and chaff, and hulled six-row barley grains and two domesticated pulses, lentil and Pisum sp. (hereafter pea). There are four tree/shrub species, caper, fig, pistachio, possibly specimens of either pear or apple and 24 wild herbaceous taxa (Hansen 2005, 232). The plant assemblage is comparative to contemporary sites with regards to the crop assemblage. However, there are considerably more wild herbaceous taxa recorded at Kalavasos-Tenta than for KissonergaMylouthkia and Parekklisha-Shillourokambos; including charred specimens of Ranunculus sp., Geranium sp., Astragalus sp. Genista sp., Medicago sp., Trifolium sp., Rubus sp., Lithospermum arvensis (hereafter field gromwell), Amaranthus retroflexus (hereafter redroot pigweed), Eleocharis sp., Schoenus nigricans (hereafter black bogrush), Scirpus sp., Hordeum murinum (wild barley), and Phalaris canariensis (hereafter canary grass). This possibly could be explained by the fact that the number of wild taxa species are lowest in the initial stages of colonisation (Colledge et al. 2004). Hansen interprets the assemblage as evidence of cereal and legume agriculture supplemented with the gathering of wild resources. Einkorn wheat is more common than emmer wheat in Period 4 (2% of total seeds versus 0.9%, correspondingly) and emmer wheat is more abundant than einkorn in Periods 2 and 3 (1% versus 0.7% total seeds). However, more evidence is needed in support of these chronology changes (Hansen 2005, 326).

KHOLETRIA-ORTOS Kholetria-Ortos is located 20 km east of Paphos, western Cyprus. Six radiocarbon dates give a range between 6600 and 5000 cal. BC, with an average date ca. 6200 cal. BC (based on one standard deviation). There was little evidence for architecture at the site but ashy cultural deposits were abundant (Simmons 1994). Archaeobotanical samples were taken during the 19931994 excavation seasons. Results from the preliminary analysis by Hansen have not yet been published and those presented here were made available by Simmons (pers. comm.). The plant remains were preserved in charred form and recovered by flotation. A total of 1,571.8 litres of soil was floated from 40 contexts. The plant remains include the following crops: barley grains

DHALI-AGRIDHI Dhali-Agridhi is a late Aceramic Neolithic and Ceramic Neolithic inland site located east of the Troödos massif. There are five radiocarbon dates (Stager and Walker 1974, 219) that give a calibrated range between 71004300 BC, with Aceramic Neolithic occupation corresponding to ca. 6550 cal. BC and the Ceramic Neolithic occupation to about 4,900 cal. BC. Stewart (1974) records the findings from plant material that was preserved in charred form and was recovered by flotation 35

CROPS, CULTURE, AND CONTACT and chaff, emmer and einkorn wheat grains and chaff, lentil, pea and vetch. There is also evidence of four fruit trees and six wild herbaceous taxa. Of note is the presence of Pyrus sp. (hereafter pear) and gromwell, both which have not been recorded in samples dated earlier. Although, Hansen records specimens from Kalavasos-Tenta that could not be identified to either pear or apple (Malus sp.).

Neolithic of Cyprus is 33 and the total number of litres sampled is approximately 11,115 litres. DHALI-AGRIDHI Stewart (1974) records the findings from plant material that was preserved in charred form that was recovered by flotation from the 1972 field season. Ceramic Neolithic occupation is dated to ca. 4900 cal. BC. A total of 12,090 litres of soil was floated from 109 contexts; however, it is unclear how many samples come from contexts dated to the Ceramic Neolithic. Ceramic Neolithic occupation at Dhali-Agridhi includes the presence of only two species, grape and lentil.

KHIROKITIA-VOUNOI Khirokitia-Vounoi is located 5 km from the southern coast of the island. Seventeen radiocarbon dates (Le Brun 1994, 1991) provide an average date of ca. 6150 cal. BC, based on one standard deviation. There are four published reports on archaeobotanical results from 19751990. The plant remains were preserved in charred form and recovered by flotation. Waines and Stanley-Price (1977) published the results from the 1975-1976 excavation seasons. A total of 2,036 litres from 17 samples was wet-sieved using a 1.6 mm mesh (1977, 281-283). The samples were taken from floors and building material, including walls and ceilings. In 1984 Miller reported the results from 27 flotation samples taken from inside and outside domestic structures from the 1977-1978 seasons. Miller records the sample sizes in litres for all but five samples, ca. 115 litres (Miller 1984, 183-188). The third report presents the data from 73 samples (355 litres) that were processed with bucket flotation using a 1mm mesh from the 1980-1981 and 1983 seasons (Hansen 2005, 327). Samples were taken from hearths, basins, between layers of rock, and structure floors (Hansen 1989, 235-250). The fourth report presents results from 141 samples (950 litres) taken from pisé and mud brick walls recovered from the 1986 and 1988-1990 seasons (Hansen 1994, 393-395). The domestic taxa include hulled barley grains and chaff, emmer wheat grains and chaff, both one-grained and two-grained einkorn wheat grains and chaff, tentative identification of free-threshing wheat, lentil, pea, and bitter vetch. Other taxa recorded include 21 wild herbaceous taxa, seven of which have not been recorded in earlier contexts, including Trigonella sp., Pimpinella sp., Asphodelus sp., Muscari sp., Carex sp., Bromus sp., and Setaria sp.

KANTOU-KOUPHOVOUNOS Kantou-Kouphovounos is located just north of Kantou in the Limassol district, Cyprus. There are only two radiocarbon dates, with an average date ca. 4,600 cal. BC. The site was excavated by Eleni Mantzourani (University of Athens) from 1992-1999. Of the 2.05 hectares identified in survey, only 0.09 hectares was excavated during that time. The architecture includes 39 rectilinear houses, with hearths/fireplaces, platforms, benches, grinding installations, and pits. Presence records only are reported for taxa that are represented in charred form. There are no descriptions of contextual association, sample size or retrieval methods. Margariti records the presence of wheat, barley, lentil, vetch, pea, grape, and mallow (Margariti in Mantzourani 2004). AYIOS EPIKTITOS-VRYSI Ayios Epiktitos-Vrysi is located on the northern coast of Cyprus in the Kyrenia lowlands. The site, which covers ca. 0.08 hectares, was excavated by Peltenburg from 1969-1973. Unfortunately, excavations were stopped at the beginning of 1974 due to political conflict on the island and as a result the report published in 1982 is considered to be preliminary. The site lies on uncultivated land but has been affected by erosion and development (Peltenburg 1982c). The domestic architecture consists of single-roomed rectilinear structures made of stone, pisé, and mud plaster with internal hearths. Seventeen radiocarbon dates provide evidence of three Phases of occupation (Peltenburg 1982c); early, middle and late, with a date range between 4330-4000 cal. BC and an average date of ca. 4165 cal. BC.

4.7 CERAMIC NEOLITHIC Botanical evidence for the Ceramic Neolithic of Cyprus is limited primarily to one site, Ayios Epiktitos-Vrysi, with minimal data from Dhali-Agridhi and KantouKouphovounos. The sites range in date from ca. 4900 cal. BC to ca. 4000 cal. BC. All three sites dated to the Ceramic Neolithic have macro remains that were preserved in charred form and recovered by flotation. However, Ayios Epiktitos-Vrysi is the only site with reported number of samples and volume of samples for this cultural phase and thus provides the most informative evidence for the Cypriot Ceramic Neolithic. The total number of samples reported the Ceramic

Kyllo (in Peltenburg 1982c) published the results of the poorly preserved charred macro remains from the 1972 field season. The samples were dry sieved with a 1 cm mesh and then the charred material was recovered from a modified froth flotation system using 2mm, 1mm and 0.5 mm mesh sieves. A total of 11,115 litres from 33 samples were processed from a series of floors, hearths, middens, and pit fills (Kyllo in Peltenburg 1982c, 90-93). A total of 50 taxa were represented at Ayios EpiktitosVrysi including both six-row and two-row hulled barley 36

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY grains, one and two-grained einkorn grains, emmer wheat grains, free-threshing wheat grains, rye, pea, lentil, chickpea, and grass pea. In addition, there were seven oil/fibre/tree plants and 32 wild/weed taxa, 21 of which have not previously been recorded (Appendix 7).

KISSONERGA-MYLOUTHKIA Kissonerga-Mylouthkia is a multi-phase site located 5km north of Paphos on the southwest coast of the island. There are five phases of occupation dating to the Early and Middle Chalcolithic, four of which have botanical data (second, third, fourth, and final phase) (Colledge 2003, 239-245). Nine radiocarbon dates dated to Period 2 and Period 3 give an average date of 3555 cal. BC (Peltenburg 2003, 259).

4.8 CHALCOLITHIC There are six sites with charred macro remains that were recovered by flotation dated to the Cypriot Chalcolithic. These sites are Kalavasos-Tenta, Lemba-Lakkous, Kissonerga-Mylouthkia, Kissonerga-Mosphilia, Kalavasos-Ayious, and Prastio-Agios Savvas. These sites range in date from ca. 4000 cal. BC to ca. 2300 cal. BC and are assigned culturally to the Early to Late Chalcolithic. With the exception of Lemba-Lakkous and Kalavasos-Ayious, all sites have recorded number and volume of samples. The total number of samples reported for the Cypriot Chalcolithic is 366 (exclusion of unknown samples from Kalavasos-Ayious) and the total number of litres sampled is approximately 14,869 litres (with the exclusion of data from Kalavasos-Tenta, Lemba-Lakkous and Kalavasos-Ayious).

The well-preserved charred plant remains were recovered by flotation. Colledge records the results from 9 samples from three pits (Pits 1, 16, and 28) with a total volume of 2,450 litres. The pits were rich in both material culture and plant remains and have been interpreted as representing areas of domestic waste disposal (Colledge 2003, 244). There was a positive correlation between both the number of taxa and the numbers of identifiable items and sample size, relating possibly to the nature of pit deposition and the greater likelihood of good preservation. Also, the density of economic plant resources and food processing debris has been interpreted as representing burnt remains from unintentional burning of storage contexts (Colledge 2003, 244). The domesticated plant taxa include rye grains, both free-threshing and glume wheat grains and chaff, six-row and two-row hulled barley grains and chaff, emmer wheat grains, chickpea, pea, bitter vetch, and lentil. In addition there are six shrub/oil/tree plants and 23 wild herbaceous taxa.

KALAVASOS-TENTA There are eight archaeobotanical samples that were taken from the Chalcolithic occupation at Kalavasos-Tenta (Hansen 2005). However, Todd (2005) cautions acceptance of the dates for this period due to possible contamination. Nevertheless, the poorly preserved charred macro remains were recovered by flotation and analysed by Hansen along with the results from Aceramic Neolithic occupation (2005). Unlike information recorded for the Aceramic Neolithic, no information regarding context or sample sizes is provided. The samples from this period contained no evidence of cereal crops only a few wild herbaceous taxa including, mallow, fumitory, and gromwell (Hansen 2005, 328).

KISSONERGA-MOSPHILIA Kissonerga-Mosphilia is a multi-phase site in Kissonerga village located 5km north of Paphos on the southwest coast of the island. The site has evidence of occupation from the Aceramic Neolithic to the Early Bronze Age. Thirty radiocarbon dates from Periods 2-5 were used to define an Early, Middle, and Late Chalcolithic occupation. The radiocarbon dates give a date range between 3500-2300 cal. BC and an average date of ca. 2900 cal. BC (Peltenburg 1998).

LEMBA-LAKKOUS Lemba-Lakkous is located in Lemba village on the southwest coast of the island. Nine radiocarbon dates (Peltenburg 1985) range in date from ca. 4,700 cal. BC to ca. 2400 cal. BC and give an average date of ca. 3800 cal. BC. Colledge reports the botanical results that were poorly preserved in charred form and recovered by flotation from the 1976-1983 excavations (Colledge 1985, 297-298). Fifteen samples from burial contexts, pits, occupation levels, and fire pits were taken from Area II, which had more charred plant remains. However, the volume of the majority of samples is unrecorded. Noteworthy is the abundance of barley and the large amount of economic plant taxa recovered from burial contexts (Colledge 1985, 297). There are a total of 20 taxa including six-row hulled barley grains, freethreshing wheat grains, indeterminate freethreshing/glume wheat grains, lentil, grape, pistachio, olive, fig and 11 wild herbaceous taxa.

Murray extensively reports the finds from the Chalcolithic occupation and her interpretations are summarised here (Murray 1998, 215-223). The botanical remains were preserved through charring and recovered by flotation using both a 1mm and 250 micron mesh. Samples were taken from 18 contexts: from pits, paved and unpaved surfaces, floors, hearths, ovens, graves, pot spreads, and general levels. A total of 10,881 litres of soil was processed from 306 samples, 248 of which could be confidently assigned to the Aceramic Neolithic (Period 1A), Early Chalcolithic (Period 2), Middle Chalcolithic (Period 3A and 3B), Late Chalcolithic (Period 4), and the Early Bronze Age (Philia, Period 5); however, due to ploughing and soil disturbance preservation of the charred material was poor and only one seed was recovered from Period 1A and only 10 seeds were recovered from the Period 5, Philia phase. 37

CROPS, CULTURE, AND CONTACT Sixteen samples were taken from Phase 2 (Early Chalcolithic), 24 samples were taken from Phase 3A (Middle Chalcolithic), and 55 samples were taken from Phase 3B (Middle Chalcolithic). Although extensive flotation efforts are reported for all phases of occupation, the bulk of the botanical evidence comes from the 150 samples taken from Period 4, the Late Chalcolithic occupation (Murray 1998, 215-221). The economic species include emmer wheat grains and chaff, possibly einkorn wheat, free threshing wheat grains and chaff, both two-row and six-row hulled barley grains and chaff, naked barley chaff, rye, lentils, peas, chick peas, and possibly vetch and grass pea, olive, grape, pistachio, fig, hackberry, juniper, linseed/flax and caper (Murray 1998, 217). In addition, there are 63 wild herbaceous taxa, which represent 35% of the assemblage. Of note is the fact that 29 of the wild taxa do not appear in the assemblages of sites dated to earlier periods. Most of the wild herbaceous taxa are either weeds of cultivated crops or are invaders that are commonly found in cultivated areas. From the wild taxa, Murray interprets that the crops would have been sown, with a sickle blade, in the winter and harvested in the spring. Three percent of the wild herbaceous taxa are wet loving species, including species in the Cyperaceae family; possibly suggesting a shift in field location by the Late Chalcolithic (Period 4), more specifically a shift away from settlements and closer to a water resource (i.e. spring or streams and in this case near to Skotinis stream, adjacent to the site today). There is evidence of animal consumption of barley grains and other cereal crop processing wastes and evidence for the burning of animal dung burned for fuel, which subsequently was swept into the pits (Murray 1998, 220).

KALAVASOS-AYIOUS Kalavasos-Ayious is an Early Chalcolithic site located in the Larnaca district. Four radiocarbon dates (Todd and Croft 2004) range between 4,040 cal. BC and 3630 cal. BC, with an average date of ca. 3835 cal. BC. The botanical remains were preserved in charred form and recovered by flotation. Hansen records disappointing results due to poor preservation. The samples were taken from Pits 25 and 27 in the northwest area of the site. Hansen reports no information on sample size, quantities or densities but records the following taxa: emmer wheat, barley and lentil (Todd and Croft 2004). PRASTIO-AGIOS SAVVAS TIS KARONIS MONASTERY Prastio-Agios Savvas Tis Karonis Monastery is a Middle Chalcolithic (c. 3500-2800 BC) site located in the vicinity of the ruined monastery in the deserted village of Prastion, southern Cyprus (Rupp 2000). The plant remains were preserved in charred form and recovered by flotation. A total of 1,538 litres from 28 samples were processed (Murray in Rupp 2000). The plant remains include emmer wheat, free-threshing wheat, hulled barley, lentil, fig, grape and pistachio along with 11 wild herbaceous taxa associated with the cultivation of cereal and pulse crops. 4.9 BRONZE AGE OCCUPATION There are four sites dated to the Early and Middle Cypriot Bronze Age with records of botanical data. The sites are Marki-Alonia, Sotira-Kaminoudhia, EpiskopiPhaneromeni, and the Middle Bronze Age Cemetery in Kalavasos Village. Based on radiocarbon dates from Marki-Alonia and Sotira-Kaminoudhia the date range is from ca. 2400 cal. BC to ca. 1700 cal. BC. The total number of samples recorded for the Early and Middle Bronze Age is 133 and the total volume recorded is approximately 549 litres.

Similar to the Chalcolithic Kissonerga-Mylouthkia pits, the pit contexts sampled from Kissonerga-Mosphilia also contained the highest concentration of items per litre and greatest range of taxa, which is likely due to better preservation. Murray reports differences between the two main areas of the site, the Main Area and the Upper Terrace, with the latter containing higher densities of taxa likely as a result of preservation and location on site rather than a greater intensity of agricultural activity (Murray 1998, p. 219). Of note is the Pithos House (Building 3) dated to Period 4, which contained dozens of large volume storage containers and a possible olive press. This house has been interpreted as possibly the largest, long-term storage facility in Cyprus dating before the Bronze Age. However, very few botanical specimens were recovered from this building and the pithoi within it. The possibility of a liquid being stored has been suggested but the botanical evidence does not support the substance being olive oil (Murray 1998, 220). Additionally, eight samples were taken from the ceremonial area, however, due to preservation; the taxa are only briefly discussed, which include wheat, lentil, pistachio, grape, fig, mallow, bedstraw, gromwell, clover, and flax (Murray 1991, 72).

MARKI-ALONIA Marki-Alonia is an Early and Middle Bronze Age site located in central Cyprus, northwest of the Troödos massif. Nine radiocarbon dates (Frankel and Webb 1992) provide an average date of ca. 2125 cal. BC. The botanical remains were preserved in charred form and recovered by flotation and wet-sieving. Sorting of the material was done by hand and not with the aid of a microscope. A total of 235 litres from 52 contexts was processed from the 1991-1993 seasons (Adams and Simmons in Frankel and Webb 1996). Taxa include barley, emmer wheat, chickpea, lentil, fig, Amygdalus communis (hereafter almond), pistachio, olive, grape, and 18 wild herbaceous taxa; five of which have not previously been recorded including, Oxalis sp., Galium spurium, Anthemis sp., Picris sp. and Solanum sp.

38

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY SOTIRA-KAMINOUDHIA

remains were recovered by flotation, at two specimens were hand-picked) and two with botanical data based on impressions in pottery. Helbaek records a total of 43 impressions in pottery from Kalopsidha (1966, p. 124) and Hjelmqvist records impressions from mud brick, pottery and from plant remains that were preserved through mineralisation from Hala Sultan-Tekke (1976; 1979, 110-133). Botanical remains preserved by charring and recovered by method of flotation include those from Kalavasos-Ayios Dhimitrios, Maa-Palaeokastro, Phlamoudhi (Hansen 1989, 82-93; Miksicek 1988; Smith 2008). Two-hundred and seventy samples were taken from six seasons of excavations from Kalavasos-Ayios Dhimitrios. The samples were recovered from multiple deposit types including tombs, pithoi and pottery, and storage deposits in buildings. The occupation levels of the site are relatively close to the surface (within one meter), as a consequence the remains in these levels were poorly preserved (Hansen 1989,. 82-93).

Sotira-Kaminoudhia is an Early Bronze Age site in southern Cyprus. Nine radiocarbon dates (Swiny et al. 2003; Steel 2004) provide a calibrated date range between ca. 2460 cal. BC and ca. 2140 cal. BC, with an average date of ca. 2300 cal. BC. The plant material was preserved in charred form and recovered by flotation using both 1.5 mm and 0.5 mm sieves. Nineteen samples were taken from shallow deposits of bins and pits (Hansen in Swiny et al. 2003). Very few remains were recovered from the samples and Hansen attributes this to the depth of the deposits being less than a meter from the surface which increases exposure to negative taphonomic processes. The taxa include emmer wheat, grape, almond, olive, pistachio, and pear, as well as two wild herbaceous taxa, ryegrass and brome grass. MIDDLE BRONZE AGE CEMETERY IN KALAVASOS VILLAGE

A total of 18 litres were processed from nine (18 litres) samples taken from hearth and pit in the 1985 excavation season at Maa-Palaeokastro (Miksicek 1988). No information is recorded on sample size or recovery methods at Phlamoudhi, only that the analysis was done with a 10x magnification hand lens and without the aid of a microscope (Smith 2008). The plant remains from Apliki-Karamallos were preserved in charred form. The material was recovered from the contents of pots and baskets from a burnt house and from room fill deposits and burnt floors (Helbaek 1962, 171-186; Helbaek 1966, 119; du Plat Taylor 1952, 165). Plant remains from Salamis were preserved in charred form; however it is not clear whether the remains were recovered by flotation or hand excavation. There are two archaeobotanical reports, one from the 6th-5th centuries B.C and one from the 4th century B.C. (Hjelmqvist 1973; Renfrew 1970). Interestingly, there are nine tree/shrub taxa that are not previously recorded on Cyprus. These are Citrus medica (hereafter citron), Corylus avellana (hereafter hazelnut), Ficus sycomorus (hereafter sycamore fig), Pinus pinea (hereafter stone pine), Punica granatum/Punica sp. (hereafter pomegranate), Quercus sp. (hereafter oak), Styrax officinalis (hereafter styrax), Ziziphus lotus, and Ziziphus spina-christi (hereafter Christ’s thorn jujube).

The Middle Bronze Age Cemetery in Kalavasos Village was excluded from the database discussed in Chapter 3 due to contamination in tomb contexts. The samples were taken from general tomb fills and inside pottery vessels excavated from tombs (Todd 1986). EPISKOPI-PHANEROMENI Episkopi-Phaneromeni is a Middle Bronze Age site located about 14 km west of Limassol. In 1981 Carpenter published Hansen’s analysis of the charred material from the 1975-1978 excavations. Sixty-two samples (2 litres each) were processed and the results are described as disappointing. No information regarding contexts was provided, however, Carpenter records the presence of barley, lentil, grape, and either apple or pear (Carpenter in Swiny 1981, p. 65). 4.10 LATE BRONZE AGE AND BEYOND As discussed in Chapter 2, the Late Bronze Age of Cyprus (ca. 1650 -1050 cal. BC) differs from earlier Cypriot cultural phases including new social and economic transformations including massive population increase with the rise of urban complexes and a more interactive foreign trade network (Steel 2004). This is a point in Cypriot prehistory that Cyprus becomes truly integrated in the broader Mediterranean interaction sphere and evidence of contact and trade in the archaeobotanical record would be expected. Therefore, I have used the data from the Cypriot Middle Bronze Age sites as an ending point for detailed comparative analysis and have combined the data from all Late Bronze Age sites for general chronological comparisons between the Late Bronze Age and everything leading up to it. There are eight sites with botanical data from the Late Bronze and later phases and there are differences in retrieval techniques used in the recovery of botanical data from these sites. There are five sites for which evidence is based on charred plant remains (at three sites charred

4.11 CYPRIOT ARCHAEOBOTANICAL SUMMARY What can be summarised from the crop and weed assemblages is that the agricultural farming package typical of the Khirokitian culture was introduced to the island at ca. 8500 cal. BC and the evidence from sites located in Cyprus illustrate differences in crop and weed assemblages over time. Studies that look at large-scale chronological and regional trends in this Cypriot dataset are few. In 1991 Hansen summarized the evidence for economic plant species from the Aceramic Neolithic to the Classical period and interpreted changes in taxa as relating to changes in subsistence (Hansen 1991, 225236). Since then, there have been a number of sites with 39

CROPS, CULTURE, AND CONTACT data that have contributed to a better understanding of the prehistoric economy of Cyprus, largely for the Aceramic Neolithic and which will be discussed here. Previous research has focused primarily on the domestication status of the early cereal crops, the timing of their introduction to the island and the regional origins of the island’s colonists. Differences between sites have previously been discussed, particularly in regards to differing crop packages and cereal crop proportions (van Zeist 1981, 97; Hansen 2005, 327). Prior research on crop-based agriculture for the later cultural phases (i.e. from the Ceramic Neolithic) has focused primarily on new species introduction. This discussion includes a summary of previous interpretation with regards to the following themes; crop introduction versus indigenous crop domestication, possible origins of the Cypriot farmers, variations in crop compositions, changes of proportions of taxa over time, and estimated chronology for species introductions. However, final interpretations and conclusions, with the inclusion of the data analysed in this book will be presented in Chapter 7.

not much detail reported on the identification of wild einkorn at the site. Stager and Walker (1989, p. 206) state that the “wild einkorn determination was based on the size and compressed lateral faces of the kernel”. Data from Kissonerga-Mylouthkia (Phases 1A and 1B) demonstrates evidence of the domesticated ‘founder’ crops, grains and chaff in the early Cypriot Aceramic Neolithic, the Cypro-PPNB (Peltenburg et al. 2000, 2003). Peltenburg et al. (2000, 850) argue for crop introduction from the mainland as opposed to indigenous agricultural development during the Early PPNB in consideration of Zohary’s hypothesis of a limited number of ‘domestication events’, specifically that the founder crops would have been domesticated once or at most very few times (1996, 156). ORIGINS OF THE ISLAND’S COLONISTS The origins of the Cypriot PPNB farmers have been much discussed with crop and combined crop and wild herbaceous taxa assemblages used to infer possible regionally specific mainland origins (Hansen 2001; Willcox 2003; Colledge et al. 2004; Colledge and Conolly 2007). However, the composition of cereal crops of sites dated to the Aceramic Neolithic of Cyprus do not point to a particular area of origin but rather demonstrate similarities with the southern Levant, the mainland Levant, in general, and southeast Anatolia (Colledge and Conolly 2007, 54; Colledge et al.. 2004, S47; Murray 2003, 71; Hansen 2001, 119; Willcox 2003, 237).

CROP INTRODUCTION VERSUS LOCAL DOMESTICATION Of the three principal founder cereal crops in the origins of agriculture in the Near East, wild barley (Hordeum spontaneum) is the only one that is indigenous to Cyprus and there are no historical or present-day records for Triticum boeoticum (einkorn wheat) or Triticum dicoccoides (emmer wheat) (Christodoulou 1959; Holmboe 1914; Meikle 1985; Zohary and Hopf 2000, 16-69). The key founder pulses include wild lentil (Lens orientalis), wild pea (Pisum elatius and Pisum humile), wild chickpea (Cicer echinospermum and Cicer reticulatum), and wild bitter vetch (Vicia ervilia) (Zohary and Hopf 2000, 92-120). The pulses indigenous to Cyprus are lentil and pea (Zohary and Hopf 2000, 95105). The remaining key crop of Near Eastern agriculture discussed by Zohary and Hopf (2000) is flax (Linum bienne), which is an oil and fiber source and is indigenous to the island (Zohary and Hopf 2000, 126).

Regardless of the exact region of origin, the crop and associated weed assemblage were brought to the island during the PPNB; however, interestingly not all species represented in the mainland assemblages appear in the assemblages of the early Cypriot sites. Differences in the Aceramic Neolithic assemblages include the absence of chickpea, rye, and free-threshing wheat on the island (Hansen 2001; Willcox 2003, 237). However, there is tentative evidence of free-threshing wheat at KhirokitiaVounoi, present in one sample (out of 131; Hansen 1994) and also at Dhali-Agridhi, present in two samples (out of 99) (Steward 1974; see also Colledge and Conolly 2007). Willcox (2003, 237) states that material from the later phases of the Aceramic Neolithic, from Khirokitia and Cape Andreas-Kastros in particular, show a divergence from mainland sites with an abundance of einkorn wheat and high frequencies of rye-grass. Interestingly, van Zeist has suggested that ryegrass was possibly used as an economic resource due to its abundance at Cape Andreas-Kastros (1981, 99); however, Hansen argues against this and for its presence as crop cleaning refuse or fodder (Hansen 2005, 327).

The issue of crop introduction versus local domestication of cereal crops was of interest as early as the late seventies (Waines and Stanley-Price 1977, 284). More recently, botanical results from ParekklishaShillourokambos and Kissonerga-Mylouthkia have provided somewhat contradictory evidence, as discussed briefly above. Willcox (2001, 127) reports wild barley in the earliest phase (Phase A) at ParekklishaShillourokambos with the appearance of domestic barley by the middle phase and has suggested the domestication of barley on the island. However, more recently Willcox (2003, p. 231), argues for crop introduction based on the fact that the wild ancestors of most of the plant species are not native to Cyprus (barley being the exception). Interestingly, often overlooked evidence from DhaliAgridhi provides evidence, although problematic, of wild einkorn (identified as Triticum boeoticum var. aegilopoides; number of grains and context not provided) on the island (Stager and Walker 1989, 206). There was

CHANGES IN PROPORTIONS OF TAXA WITH TIME A change in the number of domestic and wild herbaceous taxa has been discussed more recently, with a general increase over time of both domestic and wild herbaceous taxa. Colledge and Conolly (2007) present results that demonstrate an increase in wild herbaceous taxa from the 40

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY Aceramic Neolithic to the Ceramic Neolithic of Cyprus, which could be explained by research that suggests the number of weed assemblages are lowest in the initial stages of colonisation (Colledge et al. 2004; Rösch 1998; Willerding 1986 for Central Europe; Colledge and Conolly 2007). The data not only demonstrates an increase in wild herbaceous taxa from the Aceramic Neolithic to the Ceramic Neolithic in Cyprus but from the Aceramic Neolithic to the Chalcolithic, and perhaps onwards. Figure 4.3 illustrates the number of taxa for each site over time and for the Late Bronze Age sites combined. Noted in this bar chart is a general increase over time in the number of wild herbaceous taxa and the number of domesticated crops (i.e. cereals and legumes). In particular there is a marked increase in the number of wild herbaceous taxa from the Late Chalcolithic at Kissonerga-Mosphilia, Period 4.

wheat is the most common glume wheat in all phases but decreases in incidence after the Early/Middle Chalcolithic. Free threshing wheat is rare and is mostly absent in the Aceramic Neolithic (with the exception of tentative evidence discussed above) and rare in the subsequent Ceramic Neolithic, Chalcolithic, and Bronze Age. Glume wheat is far more ubiquitous in the samples from the earlier sites, particularly one-grained einkorn and emmer wheat. All four wheats discussed (onegrained, two-grained einkorn, emmer wheat, and freethreshing wheat) are present in the Ceramic Neolithic and einkorn and free-threshing wheat decrease in the subsequent Chalcolithic. There is variation in the different author’s identifications of barley, particularly in the Aceramic Neolithic; where all levels of identification are present (i.e. intermediate hulled, two- and six- row hulled, intermediate hulled/naked, naked, and wild). Of note is the percentage of naked barley, which is infrequent in all phases. The exception is the Ceramic Neolithic but this should be interpreted with caution because of the lower number of sites dated to the Ceramic Neolithic with data. The presence of lentil and chickpea is quite common and occurs in every cultural phase; however, lentil decreases in the number of sites for which it occurs after the Chalcolithic. Bitter vetch and faba bean are rare and the latter only occurs in the Late Bronze Age (Figure 4.6).

Exceptions to this pattern of increasing taxa over time seen in Figure 4.3 can perhaps be explained by either negative taphonomic processes and/or adverse archaeobotanical retrieval methods, including sample size, flotation method (i.e. sieve size), and identification processes (i.e. identification without the aid of a microscope). For instance, exceptions for the Aceramic Neolithic include Parekklisha-Shillourokambos, DhaliAgridhi, Kholetria-Ortos, and Cape Andreas-Kastros. At Parekklisha-Shillourokambos, issues of taphonomic processes are likely the cause of the limited number of taxa as opposed to ineffective retrieval methods and sample size, which could possibility explain the limited number of taxa recovered from Dhali-Agridhi (Colledge and Conolly 2007, p 55-56), Cape Andreas-Kastros, and Kholetria-Ortos. Also for the Ceramic Neolithic, no information on sample size, context, or recovery methods is provided for Kantou-Kouphovounos. Poor preservation could explain the limited number of taxa recovered from Chalcolithic Lemba-Lakkous and Prastio-Agios Savvas (Colledge 1985; Rupp 2000). However, since all three Early/Middle Chalcolithic sites have similar numbers of wild herbaceous taxa, the possibility of a marked increase during the Late Chalcolithic (KissonergaMosphilia) raises interesting questions particularly with regards to the cultural evidence that suggests an increasing level of external contact. For Early/Middle Bronze Age Marki-Alonia, small sample size, retrieval and identification methods are all possible explanations for the relatively limited number of taxa in the plant assemblage. Also, the limited data recovered from Episkopi-Phaneromeni and Sotira-Kaminoudhia could be a result of a combination of small sample size and shallow deposits, which creates adverse preservation conditions (Hansen 2003; Carpenter 1981).

STAGGERED SPECIES INTRODUCTIONS Prior research on crop-based agriculture of the Ceramic Neolithic has focused primarily on new species introduction. Previous studies have highlighted the introduction of rye and free-threshing wheat in the Ceramic Neolithic (Hansen 1991; Colledge and Conolly 2007). In addition there is the appearance of several pulses and an increase in domestic and wild herbaceous taxa at this time (Colledge and Conolly 2007, 61). If the botanical record of the Aceramic Neolithic is truly representative of the past, new crop introductions in the later phases of the Aceramic Neolithic (Khirokitian) may include six-row hulled barley, naked barley, freethreshing wheat, chickpea, flax and the following tree species: hackberry, olive, plum, pear and grape. Ceramic Neolithic introductions may include naked barley (2-row and 6-row), grass pea, flax, and rye. Note, previous evidence of free-threshing wheat in the Aceramic Neolithic is ambiguous, as mentioned previously as is evidence of chickpea at Dhali-Agridhi (one specimen in one sample out of 99). Earlier evidence of flax is present in samples from Aceramic Neolithic KissonergaMylouthkia and Cape Andreas-Kastros; however, the flax present in the Ayios Epiktitos-Vrysi sample was identified as the domesticated variety as opposed to identification at the genus level only. The Chalcolithic and Early/Middle Bronze Age do not have any new cereal or pulse crop introductions but domesticated grape likely appears in the Chalcolithic and almond in the Early/Middle Bronze. Faba bean appears for the first time in the Late Bronze Age as well multiple tree/shrub

REPRESENTATIONS OF CROP PROPORTIONS There are differences in crop representation amongst sites dated to the same cultural phase as well as changes over time. Figures 4.4 and 4.5 are figures of bar charts that show the difference in ubiquity (i.e. percent presence) for the domesticated wheat and barley. Emmer 41

CROPS, CULTURE, AND CONTACT species as mentioned above, including citron, hazel, and pomegranate.

with the establishment of agriculture (Colledge et al. 2004; Colledge and Conolly 2007). One proposed reason for this increase is a greater investment in agricultural fields in the initial phases of colonisation perhaps as a form of risk management (Colledge et al. 2004). For the later phases it is likely that a greater area of land and cultivation in new areas would have resulted in the increase in the number of wild taxa (Colledge and Conolly 2007).

With regards to wild herbaceous taxa there is an increase over time and a number of new introductions to the record for each cultural phase, including 11 for the late Aceramic Neolithic, 18 for the Ceramic Neolithic, 32 for the Chalcolithic, four for the Early/Middle Bronze Age, and 19 for the Late Bronze Age (refer to Table 4.3). It is argued that the number of wild taxa is lowest in the initial phases of island colonization, during the CyproPPNB, an increase in diversity in later cultural phases

42

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY

Figure 4.3 Bar chart of the total number of taxa

Figure 4.4 Bar chart of the number of sites per phase with the presence of domesticated wheat

43

CROPS, CULTURE, AND CONTACT

Figure 4.5 Bar chart of the number of sites per phase with the presence of domesticated barley

Figure 4.6 Bar chart of the number of sites per phase with the presence of domesticated legumes

44

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY Date

Author

Phase

Site

1950s

(du Plat Taylor 1952)

Late Bronze Age

Apliki-Karamallos

1960s

(Helbaek 1962)

Late Bronze Age

Apliki-Karamallos

(Helbaek 1966)

Late Bronze Age

Kalopsidha

Colledge (in Peltenburg 1979a)

Chalcolithic

Kissonerga-Mylouthkia

Hansen (in Todd 1978)

Aceramic Neolithic

Kalavasos-Tenta

Hansen (in Todd 1979)

Aceramic Neolithic

Kalavasos-Tenta

(Hjelmqvist 1973)

Late Bronze Age

Salamis

Hjelmqvist (1971)

Late Bronze Age

Enkomi

Hjelmqvist (1971)

Late Bronze Age

Salamis

Hjelmqvist (1977)

Late Bronze Age

Hala Sultan Tekke

Renfrew (1970)

Late Bronze Age

Salamis

(Stewart 1974)

Aceramic Neolithic

Dhali-Agridhi

(Waines and Price 1977)

Aceramic Neolithic

Khirokitia-Vounoi

(Carpenter 1981)

Bronze Age

Episkopi-Phaneromeni

(Colledge 1985)

Chalcolithic

Lemba-Lakkous

Colledge (Peltenburg 1981)

Chalcolithic

Kissonerga-Mylouthkia

(Hansen 1986)

Middle Bronze Age

Panayia Church

(Hansen 1989)

Late Bronze Age

Ayious Dhimitrios

Kyllo (Peltenburg 1982c)

Neolithic

Ayious Epiktitos-Vrysi

Legge (Peltenburg 1982c)

Neolithic

Ayious Epiktitos-Vrysi

(Miksicek 1988)

Late Bronze Age

Maa-Palaeokastro

(Miller 1984)

Aceramic Neolithic

Khirokitia-Vounoi

(Van Zeist 1981)

Aceramic Neolithic

Cape Andreas-Kastros

(Frankel and Webb 1992)

Bronze Age

Marki-Alonia

(Frankel and Webb 1994)

Bronze Age

Marki-Alonia

(Frankel et al. 1996)

Bronze Age

Marki-Alonia

(Hansen 1991)

Aceramic Neolithic

Recent Research

(Hansen 1994)

Aceramic Neolithic

Khirokitia-Vounoi

(Le Brun 1996)

Neolithic

Le Économie de Chypre

(Murray. M A 1991)

Chalcolithic

Kissonerga-Mosphilia

Murray (Peltenburg et al. 1998)

Chalcolithic

Kissonerga-Mosphilia

(Peltenburg et al. 2003)

Chalcolithic

Kissonerga-Mylouthkia

(Colledge 2004)

Aceramic Neolithic

Neolithic Revolution

(Colledge and Conolly 2007)

Aceramic Neolithic

A Review and Synthesis

(Colledge et al. 2004)

Aceramic Neolithic

Spread of Farming

(Peltenburg et al. 2000)

Aceramic Neolithic

Colonisation

Dammann (in Smith 2008)

Late Bronze Age

Phlamoudhi

(Hansen 2001)

Aceramic Neolithic

Clues to their Origins

(Swiny et al. 2003)

Early Bronze Age

Sotira-Kaminoudhia

Hansen (Todd 2005)

Aceramic Neolithic

Kalavasos-Tenta

Hansen (Todd and Croft 2004)

Chalcolithic

Kalavasos-Ayious

(Mantzourani 2003)

Ceramic Neolithic

Kantou-Kouphovounos

Murray (Rupp et al. 2000)

Chalcolithic

Prastio-Agios Savvas

(Murray 2003)

Aceramic Neolithic

Kissonerga-Mylouthkia

(Peltenburg et al. 2001)

Aceramic Neolithic

Colonists

(Willcox 2001)

Aceramic Neolithic

Shillourokambos

(Willcox 2003)

Aceramic Neolithic

Origins

(Lucas et al. 2012)

Aceramic Neolithic

Spread of Farming

1970s

1980s

1990s

2000s

Table 4.1 Cypriot Archaeobotanical Publications List

45

CROPS, CULTURE, AND CONTACT

site Kissonerga-Mylouthkia (1A/1B) Parekklisha-Shillourokambos Kalavasos-Tenta Dhali-Agridhi Cape Andreas Kastros Kholetria-Ortos (unpublished) Khirokitia-Vounoi (total) Kalavasos-Tenta Chalcolithic Dhali-Agridhi CN Ayios Epiktitos-Vrysi Kantou-Kouphovounos Lemba-Lakkous Kissonerga-Mylouthkia Kissonerga-Mosphilia Kalavasos Ayious Prastio-Agios Savvas Marki-Alonia Sotira-Kaminoudhia Episkopi-Phaneromeni

s 12 19 416 109 23 40 258 8 33 15 9 306 28 52 19 62

l 880 2446 7764 12,090 1572 3456 11,115 2,450 10,881 10-20 l/s 1,538 235 190 124

wc yes yes yes yes yes no yes yes yes yes yes no yes no yes no

contexts yes yes yes yes no no yes yes yes yes yes yes yes no yes no

p charred im./charred charred charred charred charred charred charred charred charred charred charred charred charred charred charred charred charred charred

rc flotation flotation flotation flotation flotation flotation flotation flotation flotation ds(1cm)/flotation flotation flotation flotation flotation flotation flotation flotation flotation flotation

mesh size 1mm and 250 micron .5 mm unknown /(1mm) unknown unknown 1.6 mm, 1 mm unknown /(1mm) 2 mm, 1 mm, .5 mm unknown unknown 1mm and 250 micron 1mm and 250 micron unknown 1mm and 250 micron unknown/hand sorted 1mm and .5 mm unknown

Table 4.2 Recorded number of samples, volume of samples, preservation and recovery method, whole counts provided and contextual information provided for Cypriot sites dated to Aceramic Neolithic to Middle Bronze Age (“-“ denotes not known; “im.’ denotes impressions; “ds” denotes drysieved; “s” denotes number of samples; “l” denotes sample size in litres; “wc” denotes whole counts; “p” denotes preservation type; “r” denotes recovery method”)

46

CHAPTER 4 HISTORY OF CYPRIOT ARCHAEOBOTANY

Cypro-PPNB

Khirokitian (late AN)

Ceramic Neolithic

wild barley w/d barley d barley (indet. hulled) 2-row d barley barley hulled or naked fr.th/gl. wheat one-grained einkorn two-grained einkorn einkorn/emmer wheat d/w emmer wheat emmer wheat lentil pea bitter vetch flax caper fig

6-row hulled barley naked barley fr.th. wheat chickpea d lentil

rye grass pea d flax

pistachio plum pear/apple Adonis /Adonis dentate Ranunculus Fumaria Malva M. sylvestris/nicaensis Geranium Astragalus Genista Lathyrus Medicago Scorpiurus Trifolium Rubus Galium B. arvensis/tenuiflora Echuim Amaranthus retroflexus Beta Euphorbia helioscopia Eleocharis Schoenus nigricans Scirpus Avena Hordeum /H. murinum Lolium Phalaris /P. canariensis Rumex Polygonum Stipa .

plum pear w grape Trigonella Pimpinella Anchusa Lithospermum Setaria Teucrium Asphodelus Muscari Carex Bromus Lolium perenne

Chalcolithic

Early/Middle Bronze

Late Bronze Age broad bean

w flax

hackberry olive

Papaver dubium Malcomia Gypsophila obionica Lens orientalis Eragrostis barrelieri Medicago truncatula Scorpiurus muricatus Crucianella Galium verum Arctium lappa Calendula arvensis Centaurea Senecio Hyoscyamus Solanum nigrum Veronica Cyperus Echinaria

d grape

almond

Fumaria densiflora Brassica Neslia Sisymbrium Emex spinosa Cleome Helianthermum Spergularia Malva nicaensis Coronilla scorpioides Bifora testiculata Bupleurum/ B. subovatum Sherardia Valerianella/ V. dentate Chrysanthemum coronarium Arnebia decumbens Cuscuta Solanaceae indet. Plantago Amaranthus Beta vulgaris Chenopodium album Salsola Suaeda/ S. fruticosa Thymelaea passerine Euphorbia (peplus) Onobrychis Ornithogalum Aegilops Arrenatherum elatius Setaria/Panicum Hordeum bulbosum

Galium spurium Galium tricornutum Anthemis Solanum

citron hazelnut Christ’s thorn lotus pomegranate oak styrax Fumaria officinalis Raphanus raphanistrum Lupinus Melilotus Avena fatua Vicia cracca Rosa Coriandum sativum Calendula Carthumus /C. tenuis Onopordum Illyricum Anagallis arvensis Alkanna L. apulum/L. officinale Ajuga chamaepitys Camphorosma Chrozophora Lolium temulentum Lolium rigidum

Table 4.3 List of taxa for each cultural phase that are not recorded in previous periods (excluding identification at the family level). Common names are provide for cereals, pulses, and oil/tree/shrubs (“w” denotes wild; “d” denotes domesticated; “fr.th” denotes free-threshing wheat; “gl.” denotes glume wheat; “AN” denotes Aceramic Neolithic)

47

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS scatter gram plot that shows the correlation between the number of identifiable items and the number of cereal grains. The plot shows that there was a positive correlation between the number of identifiable items and the number of cereals. Thus, there was a greater representation of cereals in samples with larger numbers of identifiable items; however the samples with the largest number of different species/taxa were not the samples with the greatest number of items (e.g., SFN 37, 7 taxa and 126 items).

5.1 INTRODUCTION In this chapter the results of the analyses of the botanical material recovered from Krittou Marottou-‘Ais Yiorkis, Prastion-Mesorotsos, Souskiou-Laona, and KissonergaSkalia, 2005-2010, are presented. At the time of writing there were additional samples still to be analysed from the 2011 and 2012 excavation seasons from SouskiouLaona, Kissonerga-Skalia and Prastion-Mesorotsos. These sites were described in Chapter 3 and, as noted, represent a diachronic set of sites from the southwest of Cyprus. As outlined in Chapter 1 the two following questions are to be addressed in this chapter: Are there differences or similarities in the plant material between samples or context types? And what plant species are present in the samples? Unfortunately it was not possible to make comparisons of plant data between samples and context types for each site due to poor preservation and the paucity of plant remains recovered. Discussions of the possible explanations for this are presented in section 5.6. As a result of the limitations to the data recovered from these sites it was decided to address issues at a more general level and to consider agricultural practices on an island-based level. This discussion will be presented in Chapters 6 and 7 where the data that are presented in this chapter have been added to the Cypriot dataset and used to address the research questions with regards to regionalism and chronological change over time. This chapter will therefore focus strictly on presenting the results of botanical remains from each site in chronological order, followed by a discussion of plant remains recovery from archaeological sites and the possible explanations for differences between the quality and quantities of charred botanical material from sites located in Cyprus. These data are then compared to sites located in the mainland Levant.

Figure 5.1 Correlation between number of identifiable items and the number of cereals (excluding data from Unit 20N40W, which had 1,257 items and 1,230 cereals), Krittou Marottou-‘Ais Yiorkis

The majority of the plant assemblage came from one particular feature within unit 20N40W SWQ, Feature 4. Samples 28, 32, 37, 43, and 46 in Feature 4 produced the greatest densities of items, including cereals. The feature was described as an oval pit containing large quantities of chipped stone and faunal remains. The total number of identifiable plant specimens from the pit was 1,255. This accounted for 90.3% of the total number of identifiable items recovered from the entire site. Further, a total of 1,213 cereal remains were recovered from this pit. This represented 93.9% of the total number of cereals recovered from the entire site. The plant remains were likely to have been charred unintentionally and then discarded into the pit, possibly as a result of cleaning of a hearth. It is probable that the charred plant remains were the result of one or very few depositional episodes. The specimens recovered from Feature 4 were in good condition, which was probably a result of the nature of preservation of pit deposits. On the basis of comparative data from pit contexts it has been shown that pit contexts protect the specimens from destructive processes including trampling. Accordingly, the plant remains preserve with less fragmentation than if they were deposited in habitation areas (e.g. building floors) (Colledge 2003, 244). A discussion of preservation of different context types sampled in Cyprus will be presented in more detail below.

5.2 KRITTOU MAROTTOU-‘AIS YIORKIS THE SAMPLES AND CONTEXT TYPES A total of 42 samples were processed from pit fills and surfaces of circular platform structures at Krittou Marottou-‘Ais Yiorkis. The volume of each sample, the number of charred items per sample, the number of taxa per sample, the number of items per litre, and the number of cereals per litre are listed in Appendix 9. Thirty samples produced no identifiable macro-remains, the maximum number of items per litre was 4.62 and maximum number of cereals per litre was 3.43. The sample with the greatest density in both items and cereal grains was SFN 28 (i.e. sample find number assigned by site director), which was taken from the lowest fill level of a large pit, Feature 4. This sample contained 739 items and four taxa (e.g., species, genera). SFN 37, also a fill level within Feature 4, had seven taxa and this was the greatest number per sample overall. Figure 5.1 is a 48

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS THE PLANT REMAINS

variety. There were two fragments of pistachio in the assemblage but due to fragmentation and quantity, no attempt was made to identify to species.

Appendix 9 lists the taxa from samples with identifiable material from Krittou Marottou-‘Ais Yiorkis and the taxa are separated into eight units, the total numbers of plant items in each SFN are presented, together with densities per litre and ubiquities for each taxon. All units contained identifiable material. A majority (57%) of the 42 samples had identifiable plant remains, which is equivalent to 70% of the total volume (3,084 L) of sediments floated. A total of 17 taxa were identified in the samples that comprised charred grains and seeds, chaff, and nutshell fragments. Although the quantity of material was relatively small, the preservation of the charred specimens was fairly good and it was possible to identify most of the specimens to species, genus, or family level.

There were a total of eight wild herbaceous taxa present in the assemblage. Unfortunately not all specimens could be identified to species or genus and subsequently these were relegated to either ‘cf.’ genus or family level (i.e. Leguminosae). The wild taxa included Lolium sp. (ryegrass), Avena sp. (oat), Bolboschoenus cf. glaucus (sea clubrush), Brassica/Sinapis spp. (mustard), Stipa sp. (feather grass), Bromus sp. (brome grass), Malva sp. (mallow), and fragments identified to Leguminosae (legume family). Ryegrass, oat, brome grass, and mallow are weeds associated with cereal cultivation and are common in the Cypriot archaeobotanical record. Feather grass is commonly associated with waste and fallow lands and can also be found on dry, rocky hillsides and pastures (Meikle 1985, 1790-1793).

Cereals were the most abundant taxa in the assemblage and consisted of einkorn wheat grain and chaff and barley grains. There were a total of 1,333 whole cereal grains and 17 glume bases. Out of the 1,333 cereal grains, 922 cereal grains (69.2%) were unidentifiable to genus or species and were classified as ‘cereal indeterminate.’ There are two morphologically distinct varieties of einkorn wheat: one-grained einkorn (e.g., one grain per spikelet) and two-grained einkorn (e.g., two grains per spikelet) (Zohary and Hopf 2000). Both varieties were recorded in the samples. Two-grained einkorn was present in 100% of the units and was recorded in 44.2% of the samples. It was the most abundant species represented at the site, with a total of 328 whole grains. This represented 79.8% of the total identifiable cereal assemblage. There was only one grain of one-grained einkorn. Einkorn wheat chaff was rare at the site and there were only 17 glume bases in SFN 32. The presence of hulled barley was not as common as einkorn wheat. There were a total of 65 whole grains of hulled barley present in 11.6% of the samples. Hordeum sativum has been used here to refer to cultivated hulled barley (indeterminate 2-row / 6-row). To establish whether 6-row barley is present (and in the absence of rachis remains) it is necessary to assess the symmetry or asymmetry of the grains: if asymmetrical grains were identified the possibility of grains from both the sixrowed variety and the two-row variety would be likely; at Krittou Marottou-‘Ais Yiorkis it was not possible. It is, therefore, possible that the 65 grains were from the tworowed hulled variety.

DISCUSSION OF RESULTS, KRITTOU MAROTTOU-‘AIS YIORKIS In summary, a total of 17 taxa were identified in the samples that comprised charred grains/seeds, chaff, and nutshell fragments. The assemblage included three cereal taxa: one-grained einkorn, two-grained einkorn, and tworow hulled barley; four pulses: lentil and possibly pea, vetch and grass pea; two fruit trees: pistachio and olive; and eight wild/weed taxa most of which are associated with the cultivation of cereal and pulse crops. The charred plant assemblage was dominated by two-grained einkorn wheat. The presence of two-grained einkorn on the island at this time has contributed to discussions on the spread of agriculture to Cyprus, particularly with regards to the timing of the introduction and the possible origins of the farming populations (Lucas et al. 2012). Prior to the Cypro-PPNB period at Krittou Marottou-‘Ais Yiorkis, there is no evidence for two-grained einkorn on the island. There is limited evidence for the cereal in the subsequent Khirokitian period at Khirokitia-Vounoi and Kalavasos-Tenta and in the Ceramic Neolithic at Ayios Epiktitos-Vrysi. However the ubiquities of the twograined variety are considerably lower in comparison to the one-grained variety at these sites (0.36%, 0.01%, and 18%, respectively) (Lucas et al.2012). The evidence of two-grained einkorn at this site is suggestive of a separate importation of cereal crops to the island during the Cypro-middle to late PPNB. The first importation event has been discussed by Willcox (2003) as the result of the first wave of crop expansion from southeast Anatolia. This includes the introduction of one-grained einkorn, emmer, and barley. Colledge et al. (2004) suggests an introduction of these species from the southern Levant by considering the evidence from both cereal crops and arable weed assemblages. However, the combination of cereals present in the Krittou Marottou‘Ais Yiorkis samples suggests yet an additional importation event, but of another cereal crop (twograined einkorn wheat) and from a different region, the Syrian Middle Euphrates (Lucas et al. 2012). The

With the exception of lentil, all pulses were too poorly preserved to identify to genus and a distinction was not made between the wild and domesticated varieties since the two are morphologically similar (Zohary and Hopf 2000, 95). There were a total of ten specimens that were identified as Pisum / Vicia sp., cf. Pisum sp. (pea), Vicia / Lathyrus sp. (vetch/grass pea), Vicia sp., or Lens sp. (lentil). Two tree taxa were identified in the samples: Olea sp. (olive) and Pistacia sp. (pistachio). Since the site dates to the Aceramic Neolithic it is likely that any olive exploitation at this time was from trees of the wild 49

CROPS, CULTURE, AND CONTACT evidence contributes to discussions of the spread of farming to the island and it highlights the complexity of this transmission. This is discussed more fully in Chapter 7.

likely to include cereal grains along with larger weeds that are similar in size and as a result more difficult to separate (e.g. Bromus sp., Stipa sp., Lolium sp., and Avena sp.), which is what the evidence at this site suggests. Bogaard et al. (2005) highlight the potential problems of using fine sieved products to discuss cropsowing times due to the potential bias towards autumn sowing in sieved products (larger weed species) and spring sowing (smaller seeds) in the by-products. Figure 5.2 is a bar chart of the genera of wild arable weeds in the samples. The flowering times of each of the species from each genus are according to those given in the Flora of Cyprus (Meikle 1977, 1985). In this chart each y-axis unit per month represents the flowering time of a species in a genus that is represented at the site; for example, for the month of April there were 38 species within the six genera that flower in that month. For a table of flowering times of arable weeds recovered from Cypriot archaeological contexts and presented in this book, refer to Appendix 13. As shown, the majority of the species are early (January-March) and intermediate (April-June) flowering, which has been associated with autumn-sown crops (Bogaard et al. 2001). This adheres to the bias associated with cleaned (sieve) products, which is indicative of autumn-sowing. Further, flowering times of the arable weeds from Kissonerga-Mylouthkia 1A and 1B (Figure 5.3), another Cypro-PPNB site, illustrates a similar pattern. Thus, the weed evidence from both fine-sieved by-products and cleaned (sieved) grain from the Cypro-PPNB supports sowing of crops in the autumn. This will be dealt with in more detail and with a comparison of the all the Cypriot arable weed data in Chapters 6 and 7.

Figure 5.2 Bar chart of the flowering time of the different arable weeds from Krittou Marottou-‘Ais Yiorkis samples.

Figure 5.3 Bar chart of the flowering time of the different arable weeds from Cypro-PPNB Kissonerga-Mylouthkia samples.

5.3 PRASTION-MESOROTSOS

There was very little evidence of crop processing at the site. Based on high glume base and weed to grain ratios, low number of cereal grains per litre and high ubiquity of weed taxa, Murray (2003, 64) concludes that the botanical assemblage from the contemporary (CyproPPNB) Kissonerga-Mylouthkia 1A and 1B levels are the result of fine sieving by-products and crop processing including winnowing, sieving and hand sorting. The density of wheat grains per litre at KissonergaMylouthkia 1A and 1B is 0.1. In contrast the mean value at Krittou Marottou-‘Ais Yiorkis is 0.41. At KissonergaMylouthkia the weed taxa represented 52% of the assemblage with 100% ubiquity, although the numbers per samples were low. The wild/weed taxa present in the samples at Krittou Marottou-‘Ais Yiorkis represented 0.02% of the total assemblage and were present in 0.14% of the samples. This is in opposition to the evidence from the Kissonerga-Mylouthkia samples. It is possible, therefore, that at Krittou Marottou-‘Ais Yiorkis either crop processing (i.e. winnowing and sieving) was carried out away from the site or the remains of processing were deposited in a context that was not sampled, or the material did not survive due to depositional or postdepositional destructive processes. The plant remains recovered from Krittou Marottou-‘Ais Yiorkis are likely the result of fine sieving or cleaned (sieved) crop. This is

THE SAMPLES AND CONTEXT TYPES This section includes an analysis of a selection of samples taken from Areas V and VI at PrastionMesorotsos, both of which contained very small quantities of identifiable charred plant material. A total of 136 specimens were identified and these comprised charred grains/seeds, nutshell, and nutshell fragments. Preservation of the charred plant material was quite poor and as a result it was not possible to identify many of the specimens to species; however a majority of the specimens could be identified to either genus or family level. It was only possible to include a small proportion in this research because they were given to the author at a later stage of her research. Though excavations at the site continue and analysis of all the samples is ongoing. Presented here and included in the comparative analysis in Chapter 6 are preliminary results from 19 samples and 980 litres. THE PLANT REMAINS In Appendix 10 the samples analysed, the volume of the samples, the number of items per sample, the total number of taxa per sample, and the densities (e.g., 50

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS number of items per litre, total number of cereals per litre are presented. As well as a list of taxa from samples with identifiable material from Areas V and VI are provided. A total of 17 taxa were identified in the samples, which comprised charred grains/seeds, nutlets, and nutshell fragments. The assemblage included three domesticated cereal taxa: emmer wheat, one-grained einkorn wheat, and hulled barley: one pulse: lentil, flax, two trees: pistachio and grape, and ten wild herbaceous taxa; all of which are associated with the cultivation of cereal and pulse crops. The largest component of the assemblage was wild herbaceous taxa, of which there were 55 specimens and contributed to 40.4% of the total number of items. There were 53 specimens identified as “cereal indeterminate” and comprised 38.9% of the total number of items. The wild herbaceous taxa included Fumaria spp., Brassica alba, Malva spp., Lathyrus / Vicia sp, Galium spp., Buglossoides tenuiflora, Lolium spp., and 26 specimens that could not be identified higher than family level and subsequently were classified as “Leguminosae indeterminate”.

of samples taken from hearths was 4, general occupation levels was 3, floors was ten, pit fill was 6, and other was 2. Figure 5.6 reveals that the larger the sample, the greater the density of items. Samples from floors and pits had the largest sample volumes but hearths and pits had the greatest density of charred plant material.

Figure 5.4 Pie chart of the proportional representation (% of total number of remains) per building at Souskiou-Laona

5.4 SOUSKIOU-LAONA THE SAMPLES AND CONTEXT TYPES A total of 2,138 litres from 64 samples were analysed from Souskiou-Laona (Operations A, B and D) 39% (25 samples) of which had charred plant remains. There were 137 items and 13 taxa identified and these consisted of charred grains/seeds, chaff, and nutshell fragments. Preservation of the charred plant remains was poor and as a result it was not possible to identify all of the specimens to species. However, most could be assigned to either genus or family. Figure 5.5 Scatter gram plot that shows the relationship between the average number of items per litre for each building and the total volume of the samples, Souskiou-Laona

Context densities of the samples were measured in terms of the number of identifiable remains per litre. These are presented in Appendix 11 as well as the volume of each sample, the total number of taxa per sample, the total number of items per litre, and the total number of cereals per litre. Overall, the density of remains in the different contexts was low with the maximum number of items per litre 0.33. The context with the greatest number of taxa (seven species/genera) and the largest number of items (64) was Unit 57, which was the richest context and comprised 47% of the total number of items recovered from the site. Unit 57 was described as an ashy pit deposit below an occupation floor in Building 34. Unit 641, another ashy pit fill in Building 34 and below Unit 57, had the greatest density of items per litre. Thus, Building 34 had the greatest number of items per building and the plant remains recovered from it represented 64% of the total number of items (Figure 5.4). Figure 5.5 is a scatter gram plot that shows the relationship between the average number of items per litre and the total volume of the samples for each building. As shown there was no obvious correlation between the density of charred remains and sample size per building. However there was a correlation between the average numbers of items per litre. The total number

THE PLANT REMAINS A total of 13 taxa were identified in the samples, which comprised charred grains/seeds, chaff, and nutshell fragments (Appendix 11). The assemblage included: two domesticated cereal taxa: emmer wheat grains and chaff, and hulled barley; one pulse: lentil; flax; three fruit trees: fig, pistachio and grape; and six wild herbaceous taxa. Cereal grains and chaff are not well represented in the samples, with 12 cereal grains and one glume base. There were two whole barley grains, one of which was asymmetrical. Thus, there is a possibility of grains from both the two-row and the six-rowed variety. The largest components of the assemblage, 65 specimens, were identified as seeds/fruits/nuts of trees, which 66.2% of the total number was fig seeds. The large number of fig seeds is not necessarily representative of its importance at the site because one fig can contain hundreds of seeds and fig seeds are robust, often survive carbonization well, and are able to withstand destruction due to the digestive tracts of animals and therefore are often ubiquitous in archaeobotanical samples. Additionally, 51

CROPS, CULTURE, AND CONTACT charred fig seeds are light and buoyant and therefore have a better chance of rising to the surface during flotation and thus being recovered than heavier specimens. There were a total of 42 items from six wild/weed taxa; however, 17 specimens were identifiable only to family level (i.e. Leguminosae and Gramineae). The remaining wild/weed seeds were identified as Galium sp., Buglossoides tenuiflora, Rumex spp. and Lolium spp.; all of which are common weed species associated with the cultivation of cereal crops.

Figure 5.8 Scatter gram plot that shows the relationship between the number of items and the number of taxa classified by trench from Kissonerga-Skalia

Appendix 12 lists the volume of each sample (in litres), the number of taxa per sample, the number of items per litre, and the number of cereals per litre. Unfortunately, the contexts were not rich in charred plant remains. However, a relationship between the number of taxa and the volume of the samples is shown and Figure 5.7 is a scatter gram plot (logarithmic scale) that reveals that there is a positive correlation. Trenches B and G had the largest number of taxa and accordingly they were the trenches with the largest volume processed. A positive correlation between the number of items and the number of taxa is also shown (Figure 5.8). Trenches B and G had the largest number of taxa and identifiable specimens. In this plot fig seeds are excluded despite representing 77.3% of the total number of items for the site. This is because fig seeds are often ubiquitous in samples and interpretations of their presence can be misleading. Trench B had the highest average density of items per litre (Table 5.1). Trenches B and G include the first MC III-LC1A monumental structures to be identified in the SW of the island (Crewe 2011 pers. comm.). Trench B includes a large curvilinear mud-plastered structure (Feature 33) measuring 2.6m x 1.9m and a large enclosed courtyard. Within this trench there was a furnace-like structure which contained an ashy silt fill (Feature 33) and multiple fire pits, some lined with Red Polished IV pottery, which have been interpreted as cooking/heating areas (Crewe 2011 pers. comm.). The high densities of items recovered from samples from Trench B are likely associated with the furnace feature, which possibly functioned as a tannor (bread oven).

Figure 5.6 Scatter gram plot that shows the relationship between the average number of items per litre for context type and the total volume of the samples, Souskiou-Laona

Figure 5.7 Scatter gram plot that shows the relationship between the number of taxa and the volume of the samples classified by trench from Kissonerga-Skalia, trend line is logarithmic scale

5.5 KISSONERGA-SKALIA THE SAMPLES AND CONTEXT TYPES Ninety-two samples (total volume 2,519 L) were analysed from Kissonerga-Skalia. These samples, of which 40% (38 samples) had charred plant remains, were taken from Trenches A, B, C, D, E, H, I, and J. A total of 863 items from 20 different taxa were recovered from the samples. These consisted of charred grains/seeds and nutshell fragments. Preservation of the charred plant remains was poor and as a result it was not possible to identify all of the specimens to species, however most could be identified to either genus or family level.

Trench Number of Samples

B 14

D 3

G 13

I 4

J 2

Number of items per litre

1.9

0.175

0.133

0.043

0.085

Table 5.1 Table showing the average number of items per litre by trench, Kissonerga-Skalia Trench B

pit fill

pot fill

Surface

other

Number of items with fig seeds

269

442

2

23

Number of items without fig seeds

81

18

4

23

Table 5.2 Table showing the average number of items per context type, Kissonerga-Skalia

52

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS Table 5.2 shows the average number of items per context type for Trench B. This chart highlights that samples from pit fills have the highest number of charred plant remains, but when fig seeds are included, pot fill contents contain considerably more items. However, any interpretation of charred plant material recovered from pot fills must consider the function of the pot. In this case, the fig seeds were recovered from a broken pithos that was re-used as a hearth base, which was recovered from inside the furnace feature. It is possible that the furnace was also used to dry figs and thus the remains of charred fig seeds could be the result of unintentional charring or of discarded figs that were intentionally thrown into the fire. Although figs are often ubiquitous, so any interpretation of them should be made with caution.

common in fallow fields. The largest component of the assemblage was from trees and shrubs, with an overall total of 719 items. However 92.7% of the total numbers of seeds/fruits/nuts were from fig seeds. The second largest component is from wild herbaceous taxa. There were a total of 102 specimens from 14 wild/weed taxa. This contributed to 11.8% of the total number of identifiable items. Sixteen specimens were identifiable only to the family level only (i.e. Leguminosae and Gramineae). There were a total of 16 pulses in the assemblage, 13 of which were lentils. Cereal grains, exclusively hulled barley, were the least well represented plant in the assemblage. There were only a total of four whole grains, which represent less than 1% of the assemblage. Unfortunately there were very few cereal grains and other charred cereal plant parts identified in the samples and thus precluding the possibility of any discussion of cereal crop processing or agricultural practices. Even though cereal grains were rare in the samples, the 14 wild herbaceous taxa are suggestive of cereal and pulse cultivation. As will be discussed in the interpretations below, some of the wild herbaceous taxa in the Kissonerga-Skalia assemblage either appear for the first time in the archaeobotanical record of Cyprus or are introduced to the island in the Late Chalcolithic and Early Bronze Age.

THE PLANT REMAINS Hulled barley was the only domesticated cereal in the samples from Kissonerga-Skalia. Also identified were two pulse crops, three fruits trees, and 14 wild herbaceous taxa. Taxa from samples from Trenches B, C, and D are listed in Appendix 12 and taxa from samples from Trenches G, I, and J are also listed in Appendix 12. Hulled barley was present in 3% of samples and there were a total of four whole grains. Hordeum sativum has been used here to refer to cultivated hulled barley since it was not possible to identify any asymmetrical grains in the assemblage. Due to poor preservation and fragmentation there were an additional 21 cereal grains that were classified as ‘cereal indeterminate’. The legumes included chickpea and lentil, with a total of three and 13 seeds respectively. The tree and vine crops included fig, pistachio and grape. There were a total of 667 fig seeds and they were present in 15% of the samples. It is difficult to distinguish between the wild and domesticated varieties. However it is likely that the seeds of fig in the samples from Kissonerga-Skalia are from the domesticated variety because the tree was large part of food production in surrounding regions at this time (Zohary and Hopf 2000; 2012, 126) and possibly domesticated as early as the Neolithic (Kislev et al. 2006). There were a total of 27 nutlets of pistachio and 25 pips of grape. The majority of the grape pips were recovered from one context, Unit 76, Trench B (25 pips). The wild herbaceous taxa were composed of the following species, genera, and families: Malva spp., Ajuga spp., Galium spp., Carthamus spp., Heliotropium spp., Amaranthus retroflexus, Thymelaea cf. passerina, Euphorbia spp., Euphorbia helioscopia, Arrhenatherum elatius, Lolium spp., Rumex spp., Leguminosae and Gramineae.

Figure 5.9 Bar chart of the flowering time of the different arable weeds from Kissonerga-Skalia samples.

Figure 5.9 is a bar chart of the genera of wild arable weeds in the samples. The flowering times of each of the species from each genus are represented as presented in the Flora of Cyprus (Meikle 1977, 1985). The figure shows a slightly different distribution of flowering times than those that were presented for both Krittou Marottou‘Ais Yiorkis and Kissonerga-Mylouthkia. The distribution of flowering times of the majority of species from the Cypro-PPNB sites discussed above, occur early summer (March to June). The distribution for the species represented in the Kissonerga-Skalia samples illustrates a similar pattern but relatively more species appear to flower later in the year, i.e. in late summer (July to August). The differences noted between the Aceramic Neolithic and Bronze Age flowering times highlights the possibility of identifying changes in agricultural practices and seasonality over time. However, the Kissonerga-Skalia data is too limited at the moment to

DISCUSSION OF RESULTS, KISSONERGA-SKALIA A total of 20 taxa were identified in the samples that comprised charred grains/seeds and nutshell fragments. The assemblage included hulled barley, chickpea, lentil, fig, pistachio, grape, and 14 wild herbaceous taxa, all of which are associated with the cultivation of crops and/or 53

CROPS, CULTURE, AND CONTACT address issues of seasonality with certainty. Some of the species in the assemblage flower after the harvest of winter cereal crops. These species could have arrived on the site in ways other than as weeds of cereal cultivation. Evidence of dung has been inferred from samples that are composed of cereal crops (both products and byproducts), weed species characteristic of different stages of crop-processing, weeds with low growth heights, the presence of wild taxa that are not associated with the cultivation of cereal crops, and fig seeds (Charles 1998; Valamoti 2004; 2007). In ethnographic studies it is shown that cereal chaff, mainly glume wheat de-husking waste are often mixed with dung in the production of dung cakes (Charles 1998; Valamoti and Charles 2005). The dung cakes are then burned as fuel and as a result the plant parts (including grains/seeds and chaff) are charred. At Kissonerga-Skalia there is little evidence of crop-processing products and no evidence of cropprocessing by-products (i.e. other cereal plant parts). However, the by-products of crop-processing are less likely to survive charring and destructive taphonomic conditions than more robust grains and seeds. Interestingly, the large number of fig seeds recovered from inside the furnace feature (Feature 33) could perhaps be the result of burning of dung from animals fed figs (Valamoti 2004; Valamoti and Charles 2005). Although results from analyses of wood charcoal are not included here, the author noted that there was very little wood charcoal in the flotation samples. This also could suggest the burning of dung for fuel. The botanical data recovered from Kissonerga-Skalia are perhaps too limited at this time to address the issue of whether or not dung burning for fuel was carried out at the site.

As is the case with all archaeological data, preserved botanical material is subject to biases in deposition, preservation, and recovery. As stated by Fuller and Weber (2005, 103) only a small portion of the seeds from a site became charred, a smaller number are preserved, a smaller number survive fragmentation, and an even smaller number are retrieved in excavation and processed in flotation. Despite this, there is a remarkable amount of charred remains recovered from archaeological contexts. Generally plant remains recovered are the result of either accidental or intentional burning and are indicative of their own “immediate circumstances” (van der Veen 2007, 979; Renfrew 1973, 21). It is unsurprising that there is considerable variation reported in the quantities and densities of botanical data across contexts, sites, and regions. It could be said that there are nearly the same number of depositional conditions and circumstances of preservation as there are contexts and samples. Differences in quality (e.g., how well or otherwise specimens are preserved) and quantity between botanical assemblages from sites located in Cyprus have been shown above and also in Chapter 4. This section discusses possible explanations for differences between charred botanical data from sites located in Cyprus and these are compared to sites located in the mainland Levant.

5.6 Fruitful contexts in Cypriot Archaeobotany The poor preservation of charred plant material in Cyprus has been noted previously (Hansen 1991, 2005) and as a result many archaeologists have felt discouraged and consequently excluded archaeobotanical analysis from their research designs. However, the situation is changing and more sites have been added to the island’s prehistoric archaeobotanical record. Several sites have made substantial archaeobotanical contributions to prehistoric research on the island, particularly Kissonerga-Mylouthkia and Kissonerga-Mosphilia (Colledge 2003; Murray 1998, 2003). The quantity and quality of plant remains recovered from these sites enabled the authors to address issues regarding the spread of agriculture to the island and the economy of Chalcolithic Cyprus, more generally. One must then question the causes for the differences between data recovered from botanically rich sites and those sites which contain very few poorly preserved specimens. It is argued here that in addition to differences in recovery (i.e. sieve size) and identification (naked eye versus microscope sorting) methods (discussed in Chapter 4), differences/inadequacies in sample size, context type, and population density are possible factors that have adversely affected the Cypriot archaeobotanical record.

Figure 5.10 Scatter gram plot that shows the correlation between the number of taxa and the volume of samples (in litres) from sites located in Cyprus and the mainland Levant. JA (Jerf el Ahmar); TQ (Tell Qaramel); DJ (Dja’e); Mos (Mosphilia); Vrysi (Ayios Epiktitos-Vysri); WJ7 (Wadi el-Jilat 7); WJ13 (Wadi el-Jilat 13; MyCH (KissonergaMylouthkia Chalcolithic), KT(Kalavasos-Tenta); KS (KissonergaSkalia); PM (Prastion-Mesorotsos); SL (Souskiou-Laona)

Figure 5.11 Scatter gram plot that shows the correlation between the number of taxa and the number of items from sites located in Cyprus and the mainland Levant

54

CHAPTER 5 IDENTIFICATION OF ARCHAEOBOTANICAL MATERIAL FROM FOUR PREHISTORIC SITES IN CYPRUS Site Mylouthkia Aceramic Neolithic Mylouthkia Chalcolithic ‘Ais Yiorkis Kalavasos-Tenta Prastion-Mesorotsos Ayios Epiktitos-Vrysi Kissonerga-Mosphilia Souskiou-Laona Kissonerga-Skalia Wadi al-Hammeh 27 Iraq ed-Dubb Wadi el Jilat 6 Wadi el-Jilat 7 Wadi el-Jilat 13 Azraq 31 Tell Qaramel Jerf el Ahmar Tell'Abr Dja'de

Total Number of Samples 12 5 42 175 19 33 306 25 14 12 32 42 68 87 61 108 227 30 229

habitation areas (e.g. structure floors and general fill) have comparatively lower seed densities. The lower densities in these contexts could be a result of the inhabitants keeping the areas swept and tidy. Also, preservation of the charred remains would have been compromised due to fragmentation caused by constant trampling. Further, the sites that have richer seed densities tend to be mound sites that have contexts buried and protected by subsequent deposition. While, samples from tomb contexts, although protected from destructive processes, are less likely to have large amounts of charred macro remains because they are removed from domestic activities such as crop-processing and cooking, and other activities which may involve recurrent charring incidents.

Average Number of Items Per Liter 2.7 3.445 0.346 0.899 0.178 0.328 2.622 0.162 0.803 2.32 31.03 1.45 2.84 43.37 0.29 6.991 28.131 10.017 5.384

In consideration of both context type and type of site deposition, it is not surprising that the Cypriot dataset appears meager as there is no evidence for accumulated settlement mounds (i.e. tell sites) on the island and there is limited evidence of large-scale storage before the Late Chalcolithic (Peltenburg 1993). In addition, the population density of Cyprus in prehistory has always been argued to be quite low (Croft 1991; Clarke et al. 2007). Evidence for low population density has been inferred from site size, site distribution, and faunal data, particularly the correlation between societies with low populations and a reliance on hunting (Croft 1991). The discussion presented here is not intended to paint a bleak picture for the prospects of recovery on archaeological excavations in Cyprus. Rather, it seeks to highlight the importance and value of obtaining large sample sizes and to illustrate the potential for effective sampling strategies.

Table 5.3 Table of the average number of identifiable items per litre for sites located in Cyprus and the mainland Levant

Figure 5.10 is a scatter gram plot that shows the relationship (i.e. positive correlation) between the number of taxa and the volume of samples from sites located in Cyprus and the mainland Levant. The figure reveals the relationship (i.e. positive correlation) between the number of taxa and the total volume of samples for each site. The exceptions to this were sites with smaller sample sizes but with comparatively large numbers of taxa. These sites included Jerf el Ahmar and Tell Qaramel (Willcox 2002, 2012), discussed below. Figure 5.11 is a scatter gram plot that indicates a positive correlation between the number of items and the number of taxa. The plot reveals that the number of items is proportional to the number of taxa, thus the more specimens recovered, the more taxa identified. However the likelihood of identifying new species is reduced as more species are recognized because there is a limit to the number of taxa in the botanical record. A consideration of context types is necessary as well. The sites that had relatively smaller volume sizes but larger numbers of identified items, identifiable items per litre, and a greater number of taxa, such as Jerf el Ahmar, Tell Qaramel, Dja’de, Tell’Abr (Willcox 2002, 2012), Wadi el-Jilat 13, and Iraq ed-Dubb (Colledge 2001) have greater densities of plant material because they come from context types that have better preservation conditions (Table 5.3 for site densities). As presented above, there are various context types that are all subject to different deposition and preservation biases, some of which may provide protection against adverse taphonomic conditions. Contexts that tend to protect charred macro remains from destruction are usually located away from common habitation areas (Colledge 2001), such as pits, both rubbish and storage, and storage containers in general. Thus, the sites that were rich in botanical material are sites that have samples primarily from burnt storage structures (e.g. Tell’Abr), storage contexts, and pits, both rubbish and storage containers used secondarily for rubbish (e.g. Dja’de, KissonergaMylouthkia, Kissonerga-Mosphilia). In opposition, the sites where samples have been taken from primarily

5.7 CONCLUSIONS This chapter presented the results of the analyses of the botanical material recovered from Krittou Marottou-‘Ais Yiorkis, Prastion-Mesorotsos, Souskiou-Laona, and Kissonerga-Skalia. For each site the plant species present in samples were described and when possible comparisons between samples and/or context types were discussed. As a result of the limitations to the data recovered from these sites it was decided to address issues regarding agricultural practices on an island-based level. Also, the data presented in this chapter have been added to the Cypriot dataset and will be included in the multivariate comparative analysis of data from the island and the mainland Levant, the results of which will be presented in the following chapter.

55

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS 6.1 INTRODUCTION Having presented the results of the analyses of the botanical remains from four diachronic sites on Cyprus, the results from a comparative analysis of the botanical assemblages from Cypriot prehistoric sites and contemporary sites located in the mainland Levant will be presented. Archaeobotanical data from Turkey, Syria, Jordan, Egypt, Israel and Palestine, dated to the Aceramic Neolithic (hereafter AN), Ceramic Neolithic (hereafter CN), Chalcolithic (hereafter CHAL), and early and middle Bronze Age (hereafter BA) are included in this discussion. The objective of this analysis is to define any patterns in the data that might illustrate regional and chronological similarities and/or differences between sites located in Cyprus and the surrounding regions over time. Correspondence Analysis (hereafter CA) was used to explore the relationships between the Cypriot and mainland botanical data. CANACO (Ter Braak 1988) was the software program used for analyses and CANODRAW (Smilauer 1992) was the software program used to illustrate the results of CA in graphical form.

Figure 6.1 The data model (after Colledge et al. 2004 p. S37) (“*” denotes primary field; arrows show one-to-to many relationships)

All taxa identified to family level (e.g. Leguminosae, Compositae, etc.) were excluded. Included in the analysis were specimens identified to species and genus level only (following Colledge et al. 2004). Since the Cypriot dataset did not include wood charcoal data it was decided to exclude wood charcoal from the dataset for consistency. Given that the aims of the analyses were to explore similarities/differences in taxonomic composition and not individual plant parts, all categories (e.g., of grains, chaff, etc.) were combined to prevent duplication of species, and genera. Einkorn wheat and emmer wheat grains and chaff were combined to all represent glume wheat. Similarly, all specimens identified as 2-row, 6-row, or indeterminate hulled barley were pooled to represent one category of hulled barley as distinct from naked barley. This was done because of the identification problems associated with hulled barley; specifically, there have been differences and inconsistencies in the identifications of the 2-row and 6row varieties. Specimens identified as indeterminate between free-threshing wheat and glume wheat, hulled and naked barley, and rye and wheat were excluded from the analysis. The Cypriot and mainland analyses includes presence and absence data only, as a number of the authors did not report absolute quantities and at the regional level presence and absence data helps to eliminate differences between sites, contexts, and preservation of plant material. For the exploration of the wild/weed taxa, only potential weeds associated with the cultivation of crops (i.e. arable weeds) were included. This was to explore changes in agricultural practices over time. Taxa that were excluded from the analysis of the wild taxa included wild species of trees, vines, and shrubs (e.g. Olea sp., Vitis sp., Pistacia spp., Ficus sp.).

The datasets used in the analyses will be discussed, including details of the modifications made to the data, e.g. exclusions from the analyses. The first section of this chapter presents the results of the comparative analysis between Cyprus and the mainland Levant, Turkey, and Egypt. The objective is to explore the evidence for change in island contacts over time. The results of a comparative analysis of data from sites located in Cyprus are presented in the second section, which explores the relationship between samples from sites dated to the AN, CN, CHAL, and BA (early, middle, and late). The objective of this section of the analysis was to define any patterns in the data that might show chronological change in the island’s economy over time and more specifically to highlight differences in the taxonomic compositions that could suggest changes in crop choice and agricultural practices, including intensification and diversification. 6.2 REFINING THE DATASET The data used in this analysis comprise an amalgamation of three relational databases; the author’s Cypriot database, the Near Eastern database by Shennan and Conolly (2007), and The Archaeobotanical Database at the Insitute of Pre- and Protohistory and Medieval Archaeology at the University of Tübingen (Riehl and Kümmel 2005). The design of all three relational databases followed the template outlined in Colledge et al. (2004) (Figure 6.1). There were slight differences between individual recording styles and as a result it was important; first to refine the data that would be imported into the program CANACO for CA.

The sites were categorised according to the following geographical regions: Cyprus, Jordan, Syria (Euphrates Valley and central steppe), Syria (northwest, Damascus basin, and the Mediterranean coast), Turkey (central Anatolia), Turkey (southeast), Israel and Palestine Authority, and Egypt. However, for easier reading the regions will be referred to as follows: Cyprus, Jordan, Euphrates Valley (for samples from the Syrian Euphrates Valley and central steppe), western Syria (for samples in northwest Syria, Damascus basin, and the Mediterranean coast), central Anatolia (for Turkish central Anatolian 56

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS sites), SE Turkey (for samples from southeast Turkey), Israel and Palestine, and Egypt (Figure 6.2). Refer to Appendices 14 and 15 for a complete list of the sites, phase codes, taxa, and taxon codes used in this analysis. Table 6.1 is a summary of the ubiquities of cereal crops for each region and each cultural period and is the data used in the univariate analysis presented in this chapter.

6.3 COMPARATIVE ANALYSIS EXPLORATION OF THE DATASET This section presents the results of a comparative analysis of data from sites located in Cyprus, Syria, Turkey, Egypt, Israel, and Palestine and dated to the AN, CN, CHAL, and early and middle BA. The samples in the CA plots in this section were classified according to region with each country represented by a different shade and each region represented by a different symbol. Figure 6.3 is an exploratory samples plot of all samples and includes domesticated cereals and arable weed taxa. Taxa excluded from this plot include both wild and domesticated trees, vines, and legumes. Rare taxa that were present in less than 10% of the samples were excluded, as were samples (e.g. phases of sites) that contained less than five species/genera (i.e. taxon codes). These cut-off points (of rare taxa and small samples) have been successful in revealing patterns in the relationships between samples and taxa in similar CA archaeobotanical assemblages (Colledge et al 2004). The inclusion of both rare taxa and sites with small samples likely create noise and obscure patterns in the dataset (Lange 1990, 75-76, see also Jones 1991, 68-69, Colledge 2001, 183-191, Colledge et al. 2004). The total number of samples included in this plot was 169 and the total number of taxa (i.e. species/genera) was 41. The first axis accounts for 9.2% of the variation and it is along this axis that a majority of the samples from Syria are separated from the other sites. The majority of the Syrian samples have negative values on axis 1 and the majority of samples from Cyprus, Jordan, Turkey and Egypt have positive values. This exploratory plot illustrates the potential for regional patterning in the dataset. Further, it demonstrates that there are patterns in the data that are regionally as opposed to chronologically based. It highlights possible continuity in the taxonomic compositions of each region over time.

Figure 6.2 Map showing the geographical regions compared in Correspondence Analysis

Regions

Periods

N

Cyprus

Neolithic

9

56

Fr. thr. wht -*

Chalcolithic Bronze Age Neolithic Chalcolithic Bronze Age Neolithic

8 4 9 6 6 8

63 75 67 33 67 75

63 33 83 33 88

Chalcolithic

9

44

56

78

0

Bronze Age

32

78

75

81

-

Neolithic

7

100

71

100

57

Chalcolithic

1

100

100

100

0

Bronze Age

6

83

75

100

-

Neolithic Chalcolithic Bronze Age Neolithic

7 5 6 8

57 60 67 100

29 37 67 38

43 40 67 25

13 20 6 13

Chalcolithic Bronze Age Neolithic Chalcolithic Bronze Age Neolithic Chalcolithic Bronze Age

12 17 11 24 8 6

58 71 0 29 25 83

50 71 9 29 25 -

58 65 36 38 63 67

8 6 0 4 13 -

Jordan

Syria: Euphrates valley, central steppe Syria: NW, Med. coast Damascus basin Turkey: central Anatolia Turkey: southeast Israel and Palestine Authority Egypt

Gl. wht

Hulled barley

Naked barley

56

-*

100 25 67 83 50 75

0 22 0 75

EXPLANATION OF CHRONOLOGICAL GROUPINGS Patterns in the data are more clearly illustrated if the data are separated into smaller chronological groups. In the following sections the results of first, the Neolithic data, and second, both the Chalcolithic and early and middle Bronze Age data are presented. The dataset was divided into two groups based on chronology and similarities between cultural entities. The Neolithic group included the Cypro-PPNB and Khirokitian periods and the PPNB and Pottery Neolithic periods of the mainland Levant, with a date range between c. 9500 cal. BC and c. 4500 cal. BC. The Ceramic Neolithic of Cyprus was included in the CHAL and BA group. In addition to the Ceramic Neolithic of Cyprus (i.e. Ayios Epiktitos-Vrysi and limited data from Kantou), the CHAL and BA group included the CHAL and early and Middle BA of Cyprus, the mainland Levant, Turkey and Egypt, with a date range from c. 4500 cal. BC to c. 1500 cal. BC. This was due to the limited data available from the Cypriot Ceramic Neolithic and early and middle Bronze Age.

Table 6.1 Ubiquities of the cereal crops that will be used in bar charts presented in this chapter (“N” denotes number of samples/phases; “– “ denotes no data/evidence; ‘*’ denotes tentative evidence; ‘gl. wht denotes glume wheat; ‘fr. th. wht’ denotes free threshing wheat’)

57

CROPS, CULTURE, AND CONTACT However, analyses of the Chalcolithic and Bronze Age data are presented briefly to further support the patterns revealed in the CA.

left quadrant. Further, along the second axis there is a clear separation between samples located in Cyprus and those located in western Syria. The Cypriot sites have positive values on axis 2 and the Syrian sites have negative values on this axis. There also is a clear difference between the Jordanian, central Anatolian, and Syrian Euphrates samples, all of which have positive values on axis 1. It is along the second axis that a separation between samples from these regions is highlighted: the samples from sites located in central Anatolia cluster in the bottom right quadrant, the Jordanian samples primarily in the upper right quadrant, and the Euphrates Valley samples are pivotal between the two. In CANODRAW samples can be displayed as pie charts that illustrate the proportion or comparative quantities of taxa or groups of taxa present in each sample. Figure 6.5 is an equivalent pie chart plot for the results of CA portrayed in Figure 6.4. In this figure domesticated wheat, barley, and flax are represented by different shade slices, which denote the relative proportions of those taxa for each sample. This figure illustrates more clearly the relationships between the taxa and samples and facilitates interpretation of the patterns according to taxonomic composition. There is a clear separation between samples with a greater proportion of glume wheat and hulled barley and those with a greater representation of free-threshing wheat and naked barley. The samples with a greater representation of glume wheat and hulled barley are from Cyprus and western Syria and the samples with a greater representation of free-threshing wheat and naked barley are from Jordan, the Euphrates Valley and Turkey (both southeast and a majority of the central Anatolian samples). The difference between the western Syrian samples and the Cypriot samples is based on the absence of freethreshing wheat in Cyprus and the presence of freethreshing wheat and domesticated flax in Syria. The separation between the samples from Turkey is based on the presence of wild chickpea (refer to bi-plot, Figure 6.3). This is not unexpected considering the distribution of the wild progenitor species, which is located in southeast Turkey (Zohary and Hopf 2002; Weiss and Zohary 2011).

Figure 6.3 CA samples plot of arable weed taxa and domesticated cereals and flax from sites located in the Levant, Egypt, and Cyprus and dated to the AN, CN, CHAL, early and middle BA

COMPARATIVE NEOLITHIC This section presents the results of a comparative analysis of the data from the Aceramic Neolithic of Cyprus and the Aceramic and Ceramic Neolithic of the mainland Levant and Turkey. All samples in the following plots were classified according to country and region, with each country represented by a different shade and each region a different symbol. Figure 6.4 is a bi-plot of the arable weed taxa and domesticated cereals and flax. This plot shows the relationship between 49 samples and 55 species/genera after rare taxa that were present in less than 10% of the samples were excluded. The first axis accounts for 11% of the variation and the second axis represents 9.2% of the variation. This figure shows that the taxonomic compositions of the Neolithic samples were regionally distinct. There is a clear separation between the Cypriot and western Syrian samples and the central Anatolian, Euphrates Valley, and Jordanian samples. The Cypriot samples have negative values along the first axis with the samples from sites located in the western Syria. Also, hulled barley (1HULBAR) and glume wheat (1-GLWCOM) have negative values along axis 1 and are located in the top

Figure 6.6 is a bar chart that shows the ubiquity of the four cereal crops in the Neolithic. This is calculated based on the percentage of sites for each region that have evidence for each cereal crop. As shown, Cyprus is regionally distinct based on the absence of both freethreshing wheat and naked barley in the Aceramic Neolithic. All of the other regions have evidence of both glume and free-threshing wheat and both hulled and naked barley. There are regional differences in the ubiquities of cereals for each region as well. For instance, Jordan has greater ubiquities of both hulled barley and glume wheat, southeast Turkey has a higher ubiquity of glume wheat, and Syria (both regions) has more or less equal percentages of all four cereals represented. 58

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS Jordan, and Syria (both regions) have low ubiquities of chickpea, pea, and bitter vetch. Thus, the compositions of ubiquities of the domesticated cereals and legumes in Neolithic support the regional differences highlighted in the CA.

Figure 6.6 Bar chart of cereal crop ubiquities from sites located in Cyprus, Jordan, Syria, and Turkey and dated to the Neolithic period; “n” denotes number of samples

Figure 6.4 CA bi-plot of arable weed taxa and domesticated cereals and flax from sites located in the Cyprus and dated to the AN and the mainland Levant and Turkey and dated to the AN and CN

Figure 6.7 Bar chart of legume ubiquities in from Neolithic sites located in Cyprus, Jordan, Syria, and Turkey; “n” denotes number of samples

The presence of weeds in archaeobotanical studies can be used as indicators of soil fertility. In phytosociological studies (i.e. the study of the relationship between plant communities and their environments) of weeds there are two plant groups phytosociological syntaxa) that have been used as indicators of soil fertility, Chenopodietea (summer crop and ruderal weeds) and Secaliatea (winter crop weeds) (Küster 1991, 20; Wilkinson and Stevens 2003, 182-190; Jones 1999, 167; see also Zohary 1950, 387-410 for Secaliatea). Chenopodietea are typically associated with crops grown in gardens and the presence of these weeds has been used as an indicator of wellmanured gardens. In opposition, the presence of species of Secaliatea (typically grasses and perennials) has been used to infer winter-sown crops grown on less fertile soils, with little manure, and little or no irrigation (Jones 1999, 167; Bogaard et al. 2005, 505-506; Zohary 1950, 408; see also Nesbitt 2006, 9). Generally, species of the Chenopodietea and other annuals are more nitrophilous and favored by manuring and thus, have been used as a signature of intensive garden cultivation. This is in opposition to what has been inferred for the grasses and other perennial weeds, which are associated with nitrogen deficiency and little field maintenance, i.e. no

Figure 6.5 CA pie chart plot of arable weed taxa and domesticated cereals and flax from sites located in the Cyprus and dated to the AN and from sites located in the mainland Levant and Turkey and dated to the AN and CN

There are also regional differences in the compositions of domesticated legumes. Figure 6.7 is a bar chart that shows the ubiquities of domesticated legumes. Lentil is ubiquitous in all regions, although it is far less common in Jordan than the other regions. Of note are the differences in the compositions of legumes for each region, which includes a greater number of sites located in central Anatolia with chickpea as well as a greater number of sites located in Turkey (both regions) with bitter vetch and pea. With the exception of lentil, Cyprus, 59

CROPS, CULTURE, AND CONTACT soil disturbance or crop rotation (Warington 1924, 115, see also Jones 1992, 140; van der Veen 1992, 107). In this analysis, five plant families will be discussed in terms of general soil types, Chenopodiaceae, Polygonaceae, Leguminosae, Gramineae, and Cyperaceae. The plant families Chenopodiaceae and Polygonaceae are within the group Chenopodietea and will be used here as an indicator of richer, wetter, and more nitrogenous soils (i.e. well-manured) (Küster 1991, 20; Langer and Hill 1981, 197; Hanf 1983, 396-404; Grime et al. 1988, 190, 450). Also, Cyperaceae will be used as indicators of better-watered fields because species in this family grow well in moist or wet habitats (Cronquist 1981, 1139). Legumes are nitrogen-fixing, i.e. they are able to fix atmospheric nitrogen in their roots, and restore nitrogen levels in the soil (Langer and Hill 1981, 219-220; see also Grime et al. 1988, 568-574). For this reason species of the family Leguminosae can grow in nitrogen-deficient soils (Warington 1924, 119; King 1966, 150; Hanf 1983, 334). This family will be used here as an indicator of less-maintained fields and poor soils. Gramineae are within the Secaliatea group and are used as an indicator of less-fertile soils. Figure 6.8 is a samples bi-plot of only the arable weed taxa from sites dated the AN in Cyprus and the AN and CN of the mainland Levant and Turkey. This plot represents the correlation between 49 samples and 44 arable weed taxa. The first axis accounts for 11.2% of the variation and the second axis represents 10% of the variation. This plot highlights the regional patterns illustrated above and shows similar distributions, with similarities between samples from Cyprus and western Syria and parallels between samples from the Euphrates Valley and Turkey.

Figure 6.8 CA bi-plot of arable weed taxa from sites located in the Cyprus and dated to the AN and from sites located in the Levant and dated to the AN and CN

Figure 6.9 is an equivalent pie chart plot for the results of CA portrayed in Figure 6.8, which illustrates the comparative quantities of each arable weed taxa. The species were classified according to five plant families and a category of ‘other families’: Chenopodiaceae, Polygonaceae, Leguminosae, Gramineae, and Cyperaceae, with each family represented by a different shade with the exceptions of Chenopodiaceae,, Polygonaceae, and Cyperaceae, which have been assigned the same shade based on similar ecological preferences. This pie chart plot indicates that species/genera of Gramineae are ubiquitous but that there is a greater representation of species/genera that belong to the Leguminosae family in samples from Cyprus and western Syria and a greater representation of species/genera from Chenopodiaceae, Polygonaceae, and Cyperaceae in samples from Turkey and the Euphrates Valley. From this data it can be concluded that the samples from sites in Neolithic Cyprus and western Syria have poorer, less nitrogenous soils than the samples from sites located in Turkey and the Euphrates Valley; although further exploratory analyses are required to confirm the specific ecological associations.

Figure 6.9 CA pie chart plot of arable weed taxa from sites located in the Cyprus and dated to the AN and from sites located in the Levant and dated to the AN and CN

COMPARATIVE CHALCOLITHIC AND BRONZE AGE This section presents the results of a comparative analysis of the data from the Chalcolithic and early and 60

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS middle Bronze Age from sites located in Cyprus, Turkey, the Levant and Egypt. All samples in the following plots were classified according to country and region, with each country represented by a different shade and each region a different symbol. Figure 6.10 is a samples biplot of 64 samples and 52 taxa, which comprised arable weeds and domesticated cereals and flax. In this plot the locations of free-threshing wheat (FTWCOM), glume wheat (GLWCOM), and hulled barley (HULBAR) are shown. The first axis represents 13.4% of the difference and the second axis represents 7.8% of the variation. This plot shows a clear separation between the samples from all regions of Syria and those from Egypt, Cyprus, and Turkey. The Syrian samples have negative values on the first axis and the other sites have positive values along this axis. Along the second axis there is a difference between the Egyptian, Cypriot and central Anatolian samples. The Cypriot samples have positive values on axis 2, the Egyptian samples have negative values, and the central Anatolian samples are pivotal between the two. Figure 6.11 is the same bi-plot but with some of the species and sample labels that illustrate the relationship between samples and species, samples to samples, and species to species, and thus facilitates explanation of the patterning of samples in terms of taxonomic composition (i.e., the influence of crops and/or weeds on the regional groupings). This bi-plot reveals that free-threshing wheat (1-FTWCOM) is located near the point of origin and both glume wheat (1GLWCOM) and hulled barley (1-HULBAR) have positive values on the first axis and are located in the lower right quadrant.

Figure 6.10 CA bi-plot of arable weed taxa and domesticated cereals and flax from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Chalcolithic and early and middle Bronze Age

Figures 6.12 and 6.13 are bar charts that show the ubiquities of wheat and barley for the Neolithic, Chalcolithic, and early and middle Bronze Age. Ubiquity was calculated as the percentage of the number of sites for each cultural and each region with evidence for each cereal crop. What is clear from these figures is that glume wheat and hulled barley were ubiquitous in all regions, although the frequency of occurrence varies. There is an increase in the frequency of occurrence of hulled barley in the Cypriot Chalcolithic (e.g. from 56% in the Neolithic to 100% in the Chalcolithic). Also there is an increase over time in the Euphrates Valley, central Anatolia, southeast Turkey, and a slight increase in Israel and Palestine, and Egypt. Free-threshing wheat is less common than glume wheat and had the lowest ubiquity in Egypt in the Chalcolithic (25%). Naked barley is not as common as the other cereals and is more common in the samples from Neolithic, Chalcolithic, and Bronze Age Turkey; however, ubiquities are low and a slight decrease is noted for the southeast. In Syria, naked barley is common in the Neolithic and then disappears in the Chalcolithic and Bronze Age.

Figure 6.11 CA bi-plot of arable weed taxa and domesticated cereals and flax from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Chalcolithic and early and middle Bronze Age

61

CROPS, CULTURE, AND CONTACT It was suggested above that the regional differences could be based mainly on the arable weed assemblages. In the lower right quadrant of Figure 6.11 the Egyptian sites are associated more closely with the following wild taxa: Chenopodium album (4-CHENAL), Chenopodium murale (4-CHENMU), Lolium temulentum (4-LOLITE), Carex sp. (4-CARESP), and Cyperus sp. (4-CYPESP). In the case of Chenopodium album and murale, the separation of these species could be a result of differences in individual archaeobotanists’ identification criteria, particularly whether specimens were identified to the species or genus level. The CA was run on a dataset in which species of the same genus were combined and only the arable weed genera included (Figure 6.14).

Figure 6.13 Bar charts of the ubiquities of the barley crops in the Chalcolithic and early and middle Bronze Age from sites located in Cyprus, Jordan, Syria, and Turkey, Israel and Palestine Authority and Egypt. Number of Samples (‘N’ denotes Neolithic; ‘C’ denotes Chalcolithic; ‘B’ denotes Bronze Age): Cyprus: N (9), C (8), B (4); Jordan: N (9), C (6), B (6); Syria (Euphrates Valley): N= (8), C (9), B (32); Syria (western): N (7), C (1), B (6); Turkey (central Anatolia): N (7), C (5), B (6); Turkey (southeast): N (8), C (12), B (17); Israel and Palestine Authority: N (0), C (11), B (24); Egypt: N (0), C (8), B (6)

Figure 6.12 Bar charts of the ubiquities of the wheat crops in the Chalcolithic and early and middle Bronze Age from sites located in Cyprus, Jordan, Syria, and Turkey, Israel and Palestine Authority and Egypt. Number of Samples (‘N’ denotes Neolithic; ‘C’ denotes Chalcolithic; ‘B’ denotes Bronze Age): Cyprus: N (9), C (8), B (4); Jordan: N (9), C (6), B (6); Syria (Euphrates Valley): N= (8), C (9), B (32); Syria (western): N (7), C (1), B (6); Turkey (central Anatolia): N (7), C (5), B (6); Turkey (southeast): N (8), C (12), B (17); Israel and Palestine Authority: N (0), C (11), B (24); Egypt: N (0), C (8), B (6)

The bi-plot represents the results of CA on a dataset comprising 94 samples and 53 genera. The first axis accounts for 10.8% of the variation and the second axis shows 6.9% of the variation. The relationships shown in Figure 6.11 are maintained. Samples from Cyprus and central Anatolia have positive values on axis one, samples from Syria have negative values on axis one (in the lower left quadrant), and the Egyptian samples have positive values on axis one and negative values on axis two. Figure 6.14 CA bi-plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Chalcolithic, and early and middle Bronze Age

62

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS taxa and small samples) have been successful in revealing patterns in the relationships between samples and taxa as the inclusion of both rare taxa and sites with small samples likely create noise and obscure patterns in the dataset. Additional data are needed from the Chalcolithic and Bronze Age of Cyprus, in particular, to better reveal regional patterns.

Figure 6.15 CA pie chart plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Chalcolithic, and early and middle Bronze Age

Figure 6.15 is a pie-chart plot. The species are classified according to the same six plant families as in the Neolithic comparative analysis and are represented by the same shade slices (i.e. Chenopodiaceae, Polygonaceae, Leguminosae, Gramineae, and Cyperaceae) (cf. Figure 6.9). A similar pattern is displayed which shows a separation between samples from regions with perhaps more poor versus rich soils. As above, the Syrian samples are associated with a greater representation of taxa in the Leguminosae family, the Egyptian samples are associated with a greater proportion of wet ground genera in the Cyperaceae family as well as Chenopodiaceae and Polygonaceae, and the samples from Cyprus and central Anatolian have a greater representation of Leguminosae but with small proportions of Chenopodiaceae, Polygonaceae, and Cyperaceae.

Figure 6.16 CA bi-plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Chalcolithic

REGIONAL CONTINUITY IN ARABLE WEEDS Above it was shown that there were regionally distinct patterns in the compositions of arable weed taxa. In this section the sites/phases will be compared on the basis of the arable weed taxa represented in order to identify regional continuity in weed assemblages over time, from the Aceramic Neolithic to the middle Bronze Age. The samples in these plots are classified according to country and region, with each country represented by a different shade and each region a different symbol. Figure 6.18 is a sample plot that shows the relationship between 136 samples and 55 genera. Rare taxa present in less than 10% of sites/phases and samples that had less than 5 taxa were excluded. The first axis accounts for 8.3% of the variation and the second axis represents 6% of the variation. This plot illustrates the same general pattern as was shown in the CA plots of the Neolithic and Chalcolithic/Bronze Age analysis. The majority of the Syrian sites have negative values on axis one and the majority of the samples from Cyprus, Turkey, Jordon, and Egypt have positive values.

In this analysis the data from the Chalcolithic and the early and middle Bronze Age was combined due to the limited number of samples from Cyprus. The patterns illustrated in the CA revealed regional patterning on the basis of the compositions of plant taxa. When the two cultural phases are analysed separately, the same patterns are shown. Figures 6.16 and 6.17 are bi-plots based on the CA of datasets comprising arable weed genera for the Chalcolithic and early and middle Bronze Age, respectively. In the Chalcolithic period the samples from Cyprus cluster with the majority of the central Anatolian sites. While, for the early and middle Bronze Age the samples from Cyprus are associated more closely with samples from southeast Turkey. However, there was only one early and middle Bronze Age site from Cyprus (Marki-Alonia) that was included in this plot due to the cut-off points. As discussed above, cut-off points (of rare

Figure 6.19 is the pie chart plot and it clearly illustrates regional continuity over time in the representations of plant families. Thus, a clear separation between the Syrian samples and samples from Egypt is noted, the former samples have negative values on axis one (i.e. mainly in the lower left quadrant) and the latter samples have positive values on axis one and negative values on axis two (i.e. in the lower right quadrant). The first axis 63

CROPS, CULTURE, AND CONTACT can be seen to represent a scale of soil fertility. The poorer soils have negative values on axis one and wetter and richer soils have positive values on this axis. Willcox (2012) discusses the establishment of an arable weed assemblage in the Euphrates Valley during the PPNA and suggests possible continuity in this weed assemblage over time (Table 6.2 provides a list potential arable weed taxa from early Euphrates valley sites discussed by Willcox (2012) and Hillman (2000)). The evidence supports regional continuity in the arable weed assemblage of the Syrian Euphrates sites. For example, a large number of weeds highlighted by Willcox (2012) and Hillman (2000) are presented in the Syrian sites that have negative values on axis one. Figure 6.20 is a species plot in which the arable weed genera discussed by Willcox (2012) and Hillman (2000) are given different symbols to distinguish them from all other taxa. Of note also in this plot is the fact that the genera that are responsible for the separations of the Egyptian samples along the first axis are associated with moist soils (e.g. Eleocharis, Cyperus, Polygonum, Carex, Chenopodium, and Scirpus), which further supports the pattern described above.

Figure 6.18 CA samples plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Neolithic, Chalcolithic, and early and middle Bronze Age

Figure 6.17 CA bi-plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the early and middle Bronze Age

Figure 6.19 CA pie chart plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Neolithic, Chalcolithic, and early and middle Bronze Age

64

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS

Figure 6.21 Bar chart that shows the number of weed taxa in the Neolithic, Chalcolithic, and early and middle Bronze Age

SUMMARY OF COMPARATIVE ANALYSIS The results of CA highlighted regional as opposed to chronological patterning in the dataset, with continuity in the plant assemblages for each region over time. The initial exploratory plot (Figure 6.3) showed an overall separation between samples from Syria and samples from Cyprus, Jordan, Turkey and Egypt. The samples were then analysed based on cultural periods. The cultural phases were divided into two broad cultural periods: 1) Neolithic and 2) Chalcolithic and early and middle Bronze Age.

Figure 6.20 CA species plot of arable weed taxa from sites located in the Cyprus, Turkey, Egypt, and the Levant and dated to the Neolithic, Chalcolithic, and early and middle Bronze Age

taxon Adonis Aegilops Ajuga Androsace Arnebia Astragalus Atriplex Avena Bellevalia Bromus Buglossoides Bupleurum Carex Centaurea Chenopodium Coronilla Crucianella Cyperus Echinaria Echium Eleocharis Eremopyrum Euphorbia Fumaria Galium Glaucium Gypsophila Helianthemum

Willcox (2012) PPNA weeds x

*

* x * x* x x * x x

x x x* x *

taxon Heliotropium Hordeum sp. Linum Lolium Malva Medicago Melilotus Onobrychis Ornithogalum Papaver Phalaris Plantago Polygonum Prosopis Ranunculus Rumex Scirpus Scorpiurus Silene Stipa Suaeda Teucrium Thymelaea Trifolium Trigonella Vaccaria Valerianella

Exploration of the Neolithic dataset revealed regional patterns. There is a clear separation between the arable weeds and domesticated cereals of samples from Aceramic Neolithic Cyprus and those from the surrounding regions. Although differences were highlighted, Cyprus in the Aceramic Neolithic showed greater similarity with western Syria. Differences between samples from Jordan, central Anatolia, and the Euphrates Valley were also shown. The distinctions were based on both the compositions and proportions of glume wheat, free-threshing wheat, hulled barley, and naked barley. A greater representation of glume wheat and hulled barley was associated with samples from Cyprus and western Syria and a greater representation of freethreshing wheat and naked barley with the other regions (Jordan, the Euphrates Valley, and Turkey). The separation between Cyprus and western Syria was based primarily on the absence of free-threshing wheat in the Cypriot Aceramic Neolithic samples. Differences in frequency of occurrence of the cereal crops were noted with greater frequencies of both hulled barley and glume wheat noted in Jordan and a higher frequency of glume wheat in southeast Turkey. Lentils were shown to be present in all regions. There are differences in the composition of legumes in each region: in central Anatolia there are more sites with chickpea, in Turkey (both regions) there are more sites with bitter vetch and pea, and in Cyprus, Jordan, and Syria there are low frequencies of chickpea, pea, and bitter vetch.

Willcox (2012) PPNA weeds x* * * x x* x* x x * * * x * * x * x x * x* x x

Table 6.2 Table of the arable weeds in comparative analysis of all phases and all regions with those presented by Willcox 2012 denoted by “x” (these include the rare taxa that occurred at only one site). Genera listed by Willcox (2012) and not included in the analysis are Ononis, Camelina, Reseda, Convolvulus, Neslia, Isatis, Moluccella, Kickxia, and Turgenia. Also included and denoted by “*” are the potential arable weeds discussed by Gordon Hillman (Moore et al. 2000)

As above the sites/regions were also compared on the basis of the composition of the arable weeds and the analyses revealed the same regional patterns in the Neolithic data, with similarities between samples from 65

CROPS, CULTURE, AND CONTACT Cyprus and western Syria and parallels between samples from the Euphrates Valley and Turkey. The arable weeds were explored separately and showed regional variations in the proportions of five plant families (Chenopodiaceae, Polygonaceae, Leguminosae, Gramineae, and Cyperaceae). Comparisons between regions showed differences in the compositions between plant families. Gramineae is common in all regions. There was a greater representation of Leguminosae in samples from Cyprus and western Syria; and a greater representation of Chenopodiaceae, Polygonaceae, and Cyperaceae in samples from Turkey and the Euphrates Valley.

a discussion of the tree/shrub/vine data from the island and a presentation of results of an analysis of the arable weed data, with a focus primarily on the evidence for seasonality and harvesting height. CEREAL CROPS The analyses above showed that the composition of cereal taxa in Cypriot prehistory is regionally distinct. Particularly noted are the low ubiquities of free-threshing wheat and naked barley for all phases. Evidence for freethreshing wheat prior to the Ceramic Neolithic at Ayios Epiktitos-Vrysi is limited (Hansen 1991; Colledge and Conolly 2007). Equivocal evidence for free-threshing wheat in the Aceramic Neolithic comes from KhirokitiaVounoi and Dhali-Agridhi, present in 0.76% and 2% of samples, respectively (Hansen 1994, 2001; Willcox 2003, 237; Steward 1974; see also Colledge and Conolly 2007). In the Early and Middle Chalcolithic free threshing wheat is recorded at Prastio-Agios Savvas, Lemba-Lakkous, Kissonerga-Mylouthkia, and Kissonerga-Mosphilia. In the early and middle Bronze Age there is no evidence for free-threshing wheat; however, this could be due to the limited data available for this cultural period. In contrast at KissonergaMosphilia the ubiquity of free-threshing wheat declines from the early and middle Chalcolithic (Periods 2/3) to the late Chalcolithic (Period 4); from 37% ubiquity in Period 2 to 1% ubiquity in Period 4 (Murray 1998). Additional archaeobotanical data from the early and middle Bronze Age is needed to fully address the importance of free-threshing wheat at this time. It is only during the Late Bronze Age that free-threshing wheat appears to replace glume wheat. Evidence of freethreshing in the Late Bronze Age comes from Hala Sultan-Tekke, Apliki-Karamallos, Maa-Palaeokastro, and Kalopsidha. There is no evidence of glume wheat in the Late Bronze Age. The exception is a specimen at Hala Sultan-Tekke of either emmer wheat or spelt. Thus, based on current evidence, free-threshing wheat was introduced to the island either in the Khirokitian or during the Ceramic Neolithic, but did not become the more common wheat until the Late Bronze Age. Naked barley is rare in Cyprus. It is present in Neolithic samples recovered from Ayios Epiktitos-Vrysi and Dhali-Agridhi. In contrast, glume wheat is abundant in the archaeobotanical record of Cyprus (Chapter 4); in the Aceramic Neolithic both einkorn and emmer wheat were common. However, during the Chalcolithic proportional representation of einkorn decreases and the representation of emmer wheat and free-threshing wheat are similar. The numbers of sites in which hulled barley is represented increase from the Neolithic to the Chalcolithic but decrease in the early and middle Bronze Age.

The analysis of the Chalcolithic and early and middle Bronze Age data revealed similar regional separations. There are clear separations between the samples from Syria and those from Egypt, Cyprus, and Turkey. Regional differences are shown to be based more on the compositions of arable weed taxa then on the cereal crops. In particular there were differences once again in the ubiquity of free-threshing wheat and naked barley. Free-threshing wheat is represented far less frequently in Egypt and appears in both Cyprus and Egypt only in the Chalcolithic. Also, in Turkey a slight increase in the number of sites for each phase with free-threshing wheat was noted. Naked barley was not present in all regions and was more common in Turkey in the Neolithic, Chalcolithic, and Bronze Age in comparison to Cyprus, Jordan, and Israel and Palestine Authority. In Syria, naked barley is common in the Neolithic and then is absent in the Chalcolithic and Bronze Age. The sites dated to the Chalcolithic and early and middle Bronze Age were compared on the basis of the composition of the arable weeds and the analyses revealed similarities between samples from Cyprus and central Anatolia. The species were again classified according to the same five plant families as in the Neolithic comparative analysis and a similar pattern was illustrated, which showed a separation between samples from regions with perhaps more nitrogen-poor versus nitrophilous and wetter soils, the Syrian samples were associated with the poorer soils and the Egyptian samples were associated with the wetter soils. As in the Neolithic, the Syrian samples were associated with a greater representation of taxa in the Leguminosae family, the Egyptian samples with wet ground genera in the Cyperaceae family as well as Chenopodiaceae and Polygonaceae, and Cyprus and central Anatolian had a mixture of both. 6.4 CYPRUS RESULTS The results from an analysis of data from sites located in Cyprus and surrounding regions was presented above. The results showed that the taxonomic compositions of Cyprus in the Neolithic, Chalcolithic, and early and middle Bronze Age were regionally distinct. In this section the results of a comparative analysis of Cypriot data will be summarised. Also included in this section is

LEGUMES Zohary and Hopf (2000, 92) consider pea, lentil, chickpea, and bitter vetch as the key principal legumes that were taken into cultivation with the main cereal 66

CHAPTER 6 COMPARATIVE ARCHAEOBOTANICAL RESULTS crops, all four of which are present in the archaeobotanical of Aceramic Neolithic Cyprus. The number of sites with evidence for chickpea increases from the Aceramic Neolithic to the Ceramic Neolithic but then the number of sites decreases in the late Bronze Age. Although, the lack of evidence for the late Bronze Age could be due to the limited flotation efforts at excavations dated to this cultural period. In the Aceramic Neolithic, grass pea is common and then the number of sites with evidence for it decreases over time. Lentil is common in samples from all phases but its frequency decreases in the Bronze Age, again this is likely the result of lack of flotation from these sites. Faba bean (Vicia faba) is absent from the earlier cultural phases and then is introduced sometime during the Late Bronze Age.

the greatest representation are Leguminosae, Compositae, Boraginaceae, Chenopodiaceae, Liliaceae, Cyperaceae, Gramineae, and Polygonaceae. Table 6.4 lists the percentage of each of the main plant families discussed in the first section of this chapter for each cultural period. Emphasized in this chart is an increase in the proportion of Leguminosae from the Ceramic Neolithic to the Late Bronze Age, from 9.67% to 17.14%. According to a table of percentages, the proportions of Chenopodiaceae and Polygonaceae are lowest in the Late Bronze Age and greatest in the Chalcolithic. This could represent an increase in agriculture in more nitrogen poor soils during the Late Bronze Age. A discussion of this will be included in the following chapter.

TREE/SHRUB/VINES Plant Family Ranunculaceae Papaveraceae Brassicaceae Cappparaceae Cistaceae Caryophyllaceae Malvaceae Geraniaceae Oxalidaceae Leguminosae Rosaceae Cucurbitaceae Umbelliferae Rubiaceae Valerianaceae Compositae Plumbaginaceae Primulaceae Boraginaceae Convolvulaceae Solanaceae Scrophulariaceae Lamiaceae Plantaginaceae Amaranthaceae Chenopodiaceae Thymelaeaceae Euphorbiaceae Liliaceae Cyperaceae Gramineae Polygonaceae

The evidence for trees, shrubs, and vines in the Cypriot archaeobotanical record suggests a staggered introduction of species and an increase in the number of taxa over time. In the Aceramic Neolithic there is evidence for capers, fig, pistachio, plum, grape, olive, pear, hackberry, and possibly apple (identification at Khirokitia-Vounoi: Malus/Pyrus). In the early and middle Bronze Age there is the first evidence of almond. However, nine tree species first appear in the archaeobotanical records during the late Bronze Age, possibly indicating that some or all were introduced to the island at this time. The taxa that were most likely introduced include Citrus medica (hereafter citron), Punica granatum/Punica sp. (hereafter pomegranate), Ficus sycomorus (hereafter sycamore fig), and Pinus pinea (hereafter stone pine). Plant species that grow on Cyprus include Corylus avellana, Quercus sp. (hereafter oak) (hereafter hazelnut), Styrax officinalis (hereafter styrax), Ziziphus lotus, and Z. spina-christi (hereafter Christ’s thorn jujube) (Meikle 1977, 1985). ARABLE WEEDS In Chapter 4, it was shown that there is an increase in wild taxa from the Aceramic Neolithic to the Late Bronze Age (see also Colledge and Conolly 2007), particularly the most obvious increase was from the Ceramic Neolithic to the Chalcolithic. This is also noted for Jordan, Syria, and Turkey. Figure 6.21 is a bar chart that shows the number of weed taxa in the Neolithic, Chalcolithic, and early and middle Bronze Age. In all regions, with the exception of Israel and Palestine Authority and Egypt for which there are no records, there was a significant increase in the number of wild arable taxa from the Neolithic to the Chalcolithic. For Cyprus there are more arable weed genera in the samples dated to the Chalcolithic (total 64) than any other cultural phase (42 genera in the Aceramic Neolithic, 31 in the Ceramic Neolithic, 23 in the early and middle Bronze Age, and 35 in the Late Bronze Age). Table 6.3 lists the total number of taxa for each cultural phase, with each plant family represented (e.g. Leguminosae, Chenopodiaceae, etc.). As shown the plant families with

AN 2 1 1

1 1

CHAL 1 2 4 1 1 2 1

8 1

3

7

2 1

1 2

1

4

3

1

1 1

CN 2 2 1

2 1

1 2 1 3 4 8 2

1 6 2

E/M BA

LBA

1

1 3

1

1

3 1 3 1 1 3 1 1 1 2 1 1 5 1 2 3 2 11 2

1 1

6 1 2

1

1

5

4

1

1 4

1 1

1 1 1

1 1 1 1 1 4 1

1 2 1 4

Table 6.3 List of the total number of genera classified by plant family for sites dated to the Aceramic Neolithic (AN), Ceramic Neolithic (CN), Chalcolithic (CHAL), early and middle Bronze Age (E/M BA) and Late Bronze Age (LBA) of Cyprus Plant Family

AN

CN

Chalcolithic

E/M BA

LBA

Leguminosae Gramineae Cyperaceae Chenopodiaceae Polygonaceae

19.04 19.04 9.52 4.76 4.76

9.67 19.35 3.22 0 6.45

10.93 17.18 3.12 7.81 3.12

5.55 22.22 5.55 5.55 5.55

17.14 11.42 2.85 2.85 0

Table 6.4 List of the proportions of five plant families from the Aceramic Neolithic (AN), Ceramic Neolithic (CN), Chalcolithic (CHAL), early and middle Bronze Age (E/M BA) and Late Bronze Age (LBA) of Cyprus; figures are percentages

67

CROPS, CULTURE, AND CONTACT SEASONALITY

(Hillman 1985, 6). Although this is unlikely to be the case for Cyprus since glume wheats dominate. The increase in the diversity of arable taxa with low growing heights from the Aceramic Neolithic to the Bronze Age could either be the result of the greater use of freethreshing wheat over time, or be indicative of the high utility of straw in Cypriot agro-pastoral systems, i.e. the straw was an important resource for animal feed.

Knowledge of sowing times of arable weed taxa can be useful in the determination of scheduling of agricultural practices (i.e. harvesting times) as well as provide insight into general agricultural productivity, as cereal crops sown in the autumn yield more grain (Hillman 1981, 146; see also Waston et al. 1936; Kirinde 1975). Given that weeds are likely included in archaeobotanical assemblages as a result of being harvested with cereal or pulse crops, the time of harvest can be inferred based on the flowering/fruiting times of the weed taxa. For seasonality in Cypriot prehistory, the Flora of Cyprus (Meikle 1977, 1985) was used. The flowering/fruiting of each species present was recorded at the genus level and calculated on the basis of the modal value for each genus. Figure 6.22 is a bar chart that shows the distribution of the proportions of taxa (i.e. genera) that flower/fruit for each calendar month and for each cultural phase (the y-axis is labeled as the percentage of species). The figure highlights an emphasis on earlyflowering genera in all phases, with the highest percentage of genera flowering between March and May. Also noted is an increase over time in late flowering times, particularly between the months of June and September. The evidence for flowering times of the arable weed taxa suggests autumn-sowing and spring and early-summer harvesting of the crops.

Figure 6.22 Bar chart that shows the proportion of genera that flower or fruit during different months for each cultural phase in Cyprus

HARVESTING HEIGHT The growing height of modern day weeds can be used to infer the harvesting height of crops in the past (Hillman 1981, 151). In the Bandkeramik (Phases III-V) of Neolithic Europe, Kreuz et al. (2005, 249-250) noted an increase in the presence of low-growing (~ 40 cm maximum height) weeds over time and related this to changes in harvesting techniques (following Kreuz et al. (2005, 249) low harvesting height is 80 cm). For the averages of growing heights of arable weeds in Cyprus, the Flora of Cyprus (Meikle 1977, 1985) was used. The growing heights of each species present was recorded at the genus level and calculated on the basis of the modal value for each genus. Figure 6.23 is a bar chart that shows the distribution as a percentage of the species that grow at various heights for each cultural phase (refer to Appendix 13 for weed height data). A low harvesting height (i.e.,